Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,12 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-861/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-700/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoint-100/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-100/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-1.5B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.3",
+  "use_cache": true,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

checkpoint-100/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.48.3"
+}

checkpoint-100/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2de7d75467d045299ca32716e3aa556915256cd0f8d46bfea3f17351290db755
+size 3087467144

checkpoint-100/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>"
+}

checkpoint-100/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

checkpoint-100/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-100/trainer_state.json

@@ -0,0 +1,1333 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.11611030478955008,
+  "eval_steps": 500,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 724.58984375,
+      "epoch": 0.0011611030478955006,
+      "grad_norm": 0.7112110585443041,
+      "kl": 0.0009012222290039062,
+      "learning_rate": 3.846153846153846e-08,
+      "loss": 0.0,
+      "reward": 0.046875,
+      "reward_std": 0.06733439117670059,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 1
+    },
+    {
+      "completion_length": 708.56640625,
+      "epoch": 0.0023222060957910013,
+      "grad_norm": 0.12068406687948798,
+      "kl": 0.000453948974609375,
+      "learning_rate": 7.692307692307692e-08,
+      "loss": 0.0,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 2
+    },
+    {
+      "completion_length": 742.65625,
+      "epoch": 0.0034833091436865023,
+      "grad_norm": 0.7976659386050398,
+      "kl": 0.0007343292236328125,
+      "learning_rate": 1.1538461538461539e-07,
+      "loss": 0.0,
+      "reward": 0.0234375,
+      "reward_std": 0.046875,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 3
+    },
+    {
+      "completion_length": 808.21875,
+      "epoch": 0.0046444121915820025,
+      "grad_norm": 0.40112277888275455,
+      "kl": 0.00063323974609375,
+      "learning_rate": 1.5384615384615385e-07,
+      "loss": 0.0,
+      "reward": 0.060546875,
+      "reward_std": 0.1078859455883503,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 4
+    },
+    {
+      "completion_length": 654.234375,
+      "epoch": 0.005805515239477504,
+      "grad_norm": 0.39725912723012635,
+      "kl": 0.0004980564117431641,
+      "learning_rate": 1.9230769230769231e-07,
+      "loss": 0.0,
+      "reward": 0.044921875,
+      "reward_std": 0.08984375,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 5
+    },
+    {
+      "completion_length": 772.68359375,
+      "epoch": 0.006966618287373005,
+      "grad_norm": 0.14294736924830026,
+      "kl": 0.0004153251647949219,
+      "learning_rate": 2.3076923076923078e-07,
+      "loss": 0.0,
+      "reward": 0.046875,
+      "reward_std": 0.0805421955883503,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 6
+    },
+    {
+      "completion_length": 745.203125,
+      "epoch": 0.008127721335268505,
+      "grad_norm": 0.16815060965909404,
+      "kl": 0.0009188652038574219,
+      "learning_rate": 2.692307692307692e-07,
+      "loss": 0.0,
+      "reward": 0.021484375,
+      "reward_std": 0.029760945588350296,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 7
+    },
+    {
+      "completion_length": 744.265625,
+      "epoch": 0.009288824383164005,
+      "grad_norm": 12.779442066715234,
+      "kl": 0.00357818603515625,
+      "learning_rate": 3.076923076923077e-07,
+      "loss": 0.0001,
+      "reward": 0.03125,
+      "reward_std": 0.049292195588350296,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.0,
+      "step": 8
+    },
+    {
+      "completion_length": 704.51953125,
+      "epoch": 0.010449927431059507,
+      "grad_norm": 0.2475578183097461,
+      "kl": 0.0035419464111328125,
+      "learning_rate": 3.461538461538461e-07,
+      "loss": 0.0001,
+      "reward": 0.060546875,
+      "reward_std": 0.1078859455883503,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 9
+    },
+    {
+      "completion_length": 759.2890625,
+      "epoch": 0.011611030478955007,
+      "grad_norm": 0.3357083329970443,
+      "kl": 0.00310516357421875,
+      "learning_rate": 3.8461538461538463e-07,
+      "loss": 0.0001,
+      "reward": 0.00390625,
+      "reward_std": 0.0078125,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.00390625,
+      "step": 10
+    },
+    {
+      "completion_length": 746.0703125,
+      "epoch": 0.012772133526850507,
+      "grad_norm": 0.2657165785401948,
+      "kl": 0.00334930419921875,
+      "learning_rate": 4.2307692307692304e-07,
+      "loss": 0.0001,
+      "reward": 0.017578125,
+      "reward_std": 0.03515625,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 11
+    },
+    {
+      "completion_length": 702.21875,
+      "epoch": 0.01393323657474601,
+      "grad_norm": 1.2656634353615759,
+      "kl": 0.0019159317016601562,
+      "learning_rate": 4.6153846153846156e-07,
+      "loss": 0.0001,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 12
+    },
+    {
+      "completion_length": 735.78515625,
+      "epoch": 0.01509433962264151,
+      "grad_norm": 0.5357650576255997,
+      "kl": 0.005596160888671875,
+      "learning_rate": 5e-07,
+      "loss": 0.0002,
+      "reward": 0.06640625,
+      "reward_std": 0.1280700732022524,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.01171875,
+      "step": 13
+    },
+    {
+      "completion_length": 802.51953125,
+      "epoch": 0.01625544267053701,
+      "grad_norm": 1.9685839556774372,
+      "kl": 0.00266265869140625,
+      "learning_rate": 5.384615384615384e-07,
+      "loss": 0.0001,
+      "reward": 0.03125,
+      "reward_std": 0.059198048897087574,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.015625,
+      "step": 14
+    },
+    {
+      "completion_length": 702.47265625,
+      "epoch": 0.01741654571843251,
+      "grad_norm": 0.3022926592468717,
+      "kl": 0.00206756591796875,
+      "learning_rate": 5.769230769230768e-07,
+      "loss": 0.0001,
+      "reward": 0.03125,
+      "reward_std": 0.049292195588350296,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 15
+    },
+    {
+      "completion_length": 714.55859375,
+      "epoch": 0.01857764876632801,
+      "grad_norm": 0.67295700611606,
+      "kl": 0.006526947021484375,
+      "learning_rate": 6.153846153846154e-07,
+      "loss": 0.0003,
+      "reward": 0.0546875,
+      "reward_std": 0.04213257320225239,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 16
+    },
+    {
+      "completion_length": 728.94921875,
+      "epoch": 0.019738751814223514,
+      "grad_norm": 0.12683035812623522,
+      "kl": 0.003253936767578125,
+      "learning_rate": 6.538461538461538e-07,
+      "loss": 0.0001,
+      "reward": 0.041015625,
+      "reward_std": 0.0688234455883503,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 17
+    },
+    {
+      "completion_length": 752.4609375,
+      "epoch": 0.020899854862119014,
+      "grad_norm": 0.7572918382078203,
+      "kl": 0.002063751220703125,
+      "learning_rate": 6.923076923076922e-07,
+      "loss": 0.0001,
+      "reward": 0.01953125,
+      "reward_std": 0.0390625,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.00390625,
+      "step": 18
+    },
+    {
+      "completion_length": 686.39453125,
+      "epoch": 0.022060957910014514,
+      "grad_norm": 0.366311042586646,
+      "kl": 0.00254058837890625,
+      "learning_rate": 7.307692307692307e-07,
+      "loss": 0.0001,
+      "reward": 0.037109375,
+      "reward_std": 0.07421875,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 19
+    },
+    {
+      "completion_length": 814.58203125,
+      "epoch": 0.023222060957910014,
+      "grad_norm": 0.39037265038987773,
+      "kl": 0.0024509429931640625,
+      "learning_rate": 7.692307692307693e-07,
+      "loss": 0.0001,
+      "reward": 0.056640625,
+      "reward_std": 0.1000734455883503,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 20
+    },
+    {
+      "completion_length": 790.58203125,
+      "epoch": 0.024383164005805515,
+      "grad_norm": 0.16449028843577287,
+      "kl": 0.0017910003662109375,
+      "learning_rate": 8.076923076923077e-07,
+      "loss": 0.0001,
+      "reward": 0.021484375,
+      "reward_std": 0.04296875,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 21
+    },
+    {
+      "completion_length": 720.109375,
+      "epoch": 0.025544267053701015,
+      "grad_norm": 0.16050190810020273,
+      "kl": 0.000865936279296875,
+      "learning_rate": 8.461538461538461e-07,
+      "loss": 0.0,
+      "reward": 0.041015625,
+      "reward_std": 0.08203125,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 22
+    },
+    {
+      "completion_length": 772.921875,
+      "epoch": 0.026705370101596515,
+      "grad_norm": 1.022693227541973,
+      "kl": 0.010603904724121094,
+      "learning_rate": 8.846153846153846e-07,
+      "loss": 0.0004,
+      "reward": 0.0234375,
+      "reward_std": 0.046875,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 23
+    },
+    {
+      "completion_length": 785.9296875,
+      "epoch": 0.02786647314949202,
+      "grad_norm": 0.4975098781878154,
+      "kl": 0.0006268024444580078,
+      "learning_rate": 9.230769230769231e-07,
+      "loss": 0.0,
+      "reward": 0.0546875,
+      "reward_std": 0.10463257133960724,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 24
+    },
+    {
+      "completion_length": 808.91796875,
+      "epoch": 0.02902757619738752,
+      "grad_norm": 0.07636536312105723,
+      "kl": 0.0006952285766601562,
+      "learning_rate": 9.615384615384615e-07,
+      "loss": 0.0,
+      "reward": 0.03515625,
+      "reward_std": 0.055702777579426765,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.01171875,
+      "step": 25
+    },
+    {
+      "completion_length": 677.65234375,
+      "epoch": 0.03018867924528302,
+      "grad_norm": 105.92855708885655,
+      "kl": 0.048847198486328125,
+      "learning_rate": 1e-06,
+      "loss": 0.002,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.017578125,
+      "step": 26
+    },
+    {
+      "completion_length": 657.85546875,
+      "epoch": 0.03134978229317852,
+      "grad_norm": 0.0861458003290139,
+      "kl": 0.004261970520019531,
+      "learning_rate": 9.999964611162971e-07,
+      "loss": 0.0002,
+      "reward": 0.009765625,
+      "reward_std": 0.01953125,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 27
+    },
+    {
+      "completion_length": 726.2421875,
+      "epoch": 0.03251088534107402,
+      "grad_norm": 0.14527394620380304,
+      "kl": 0.00072479248046875,
+      "learning_rate": 9.999858445152839e-07,
+      "loss": 0.0,
+      "reward": 0.033203125,
+      "reward_std": 0.053198445588350296,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 28
+    },
+    {
+      "completion_length": 646.53125,
+      "epoch": 0.03367198838896952,
+      "grad_norm": 0.4038815325919152,
+      "kl": 0.008038520812988281,
+      "learning_rate": 9.999681503472433e-07,
+      "loss": 0.0003,
+      "reward": 0.02734375,
+      "reward_std": 0.043573048897087574,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.01171875,
+      "step": 29
+    },
+    {
+      "completion_length": 676.234375,
+      "epoch": 0.03483309143686502,
+      "grad_norm": 24.530460542475996,
+      "kl": 0.08559513092041016,
+      "learning_rate": 9.99943378862646e-07,
+      "loss": 0.0034,
+      "reward": 0.076171875,
+      "reward_std": 0.1358339935541153,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 30
+    },
+    {
+      "completion_length": 669.62890625,
+      "epoch": 0.035994194484760524,
+      "grad_norm": 12.948432597412872,
+      "kl": 0.0016450881958007812,
+      "learning_rate": 9.999115304121457e-07,
+      "loss": 0.0001,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 31
+    },
+    {
+      "completion_length": 632.80078125,
+      "epoch": 0.03715529753265602,
+      "grad_norm": 0.14738843577567481,
+      "kl": 0.0011644363403320312,
+      "learning_rate": 9.998726054465744e-07,
+      "loss": 0.0,
+      "reward": 0.0546875,
+      "reward_std": 0.106073047965765,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.015625,
+      "step": 32
+    },
+    {
+      "completion_length": 686.90234375,
+      "epoch": 0.038316400580551524,
+      "grad_norm": 6.995419155886516,
+      "kl": 0.056095123291015625,
+      "learning_rate": 9.998266045169354e-07,
+      "loss": 0.0022,
+      "reward": 0.052734375,
+      "reward_std": 0.0922609455883503,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 33
+    },
+    {
+      "completion_length": 655.90234375,
+      "epoch": 0.03947750362844703,
+      "grad_norm": 3.5735767009613224,
+      "kl": 0.013265609741210938,
+      "learning_rate": 9.997735282743968e-07,
+      "loss": 0.0005,
+      "reward": 0.076171875,
+      "reward_std": 0.13439351692795753,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 34
+    },
+    {
+      "completion_length": 639.82421875,
+      "epoch": 0.040638606676342524,
+      "grad_norm": 1.8475352137430245,
+      "kl": 0.0025959014892578125,
+      "learning_rate": 9.997133774702812e-07,
+      "loss": 0.0001,
+      "reward": 0.05078125,
+      "reward_std": 0.09682007133960724,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.01953125,
+      "step": 35
+    },
+    {
+      "completion_length": 715.3125,
+      "epoch": 0.04179970972423803,
+      "grad_norm": 13.632018237363075,
+      "kl": 0.21777725219726562,
+      "learning_rate": 9.996461529560552e-07,
+      "loss": 0.0087,
+      "reward": 0.0390625,
+      "reward_std": 0.078125,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.015625,
+      "step": 36
+    },
+    {
+      "completion_length": 631.8203125,
+      "epoch": 0.042960812772133525,
+      "grad_norm": 10.015816840984483,
+      "kl": 0.00942230224609375,
+      "learning_rate": 9.995718556833178e-07,
+      "loss": 0.0004,
+      "reward": 0.072265625,
+      "reward_std": 0.141229297965765,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.025390625,
+      "step": 37
+    },
+    {
+      "completion_length": 636.9140625,
+      "epoch": 0.04412191582002903,
+      "grad_norm": 0.18443450708075712,
+      "kl": 0.001720428466796875,
+      "learning_rate": 9.994904867037865e-07,
+      "loss": 0.0001,
+      "reward": 0.033203125,
+      "reward_std": 0.046594543382525444,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.017578125,
+      "step": 38
+    },
+    {
+      "completion_length": 667.6484375,
+      "epoch": 0.045283018867924525,
+      "grad_norm": 0.7286107675191874,
+      "kl": 0.027004241943359375,
+      "learning_rate": 9.994020471692832e-07,
+      "loss": 0.0011,
+      "reward": 0.0625,
+      "reward_std": 0.10044586285948753,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.0234375,
+      "step": 39
+    },
+    {
+      "completion_length": 695.68359375,
+      "epoch": 0.04644412191582003,
+      "grad_norm": 0.7697487269128109,
+      "kl": 0.002681732177734375,
+      "learning_rate": 9.993065383317162e-07,
+      "loss": 0.0001,
+      "reward": 0.078125,
+      "reward_std": 0.12407248839735985,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.0234375,
+      "step": 40
+    },
+    {
+      "completion_length": 678.390625,
+      "epoch": 0.04760522496371553,
+      "grad_norm": 0.33385332935798095,
+      "kl": 0.0028896331787109375,
+      "learning_rate": 9.992039615430648e-07,
+      "loss": 0.0001,
+      "reward": 0.06640625,
+      "reward_std": 0.1137621309608221,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.01953125,
+      "step": 41
+    },
+    {
+      "completion_length": 660.2421875,
+      "epoch": 0.04876632801161103,
+      "grad_norm": 5.2234598338711615,
+      "kl": 0.03473663330078125,
+      "learning_rate": 9.990943182553578e-07,
+      "loss": 0.0014,
+      "reward": 0.05859375,
+      "reward_std": 0.11058359593153,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.03515625,
+      "step": 42
+    },
+    {
+      "completion_length": 599.66015625,
+      "epoch": 0.04992743105950653,
+      "grad_norm": 0.3125631745813204,
+      "kl": 0.0029144287109375,
+      "learning_rate": 9.989776100206547e-07,
+      "loss": 0.0001,
+      "reward": 0.087890625,
+      "reward_std": 0.1592714935541153,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.033203125,
+      "step": 43
+    },
+    {
+      "completion_length": 695.41015625,
+      "epoch": 0.05108853410740203,
+      "grad_norm": 0.31129440427202076,
+      "kl": 0.002468109130859375,
+      "learning_rate": 9.98853838491023e-07,
+      "loss": 0.0001,
+      "reward": 0.046875,
+      "reward_std": 0.08240367192775011,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.0234375,
+      "step": 44
+    },
+    {
+      "completion_length": 594.67578125,
+      "epoch": 0.05224963715529753,
+      "grad_norm": 0.2962376329942141,
+      "kl": 0.00382232666015625,
+      "learning_rate": 9.98723005418515e-07,
+      "loss": 0.0002,
+      "reward": 0.091796875,
+      "reward_std": 0.1253724191337824,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.044921875,
+      "step": 45
+    },
+    {
+      "completion_length": 672.73046875,
+      "epoch": 0.05341074020319303,
+      "grad_norm": 3.778170631942383,
+      "kl": 0.077880859375,
+      "learning_rate": 9.985851126551428e-07,
+      "loss": 0.0031,
+      "reward": 0.103515625,
+      "reward_std": 0.1781556848436594,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.041015625,
+      "step": 46
+    },
+    {
+      "completion_length": 603.15625,
+      "epoch": 0.054571843251088534,
+      "grad_norm": 0.3069672598383264,
+      "kl": 0.006290435791015625,
+      "learning_rate": 9.98440162152852e-07,
+      "loss": 0.0003,
+      "reward": 0.08984375,
+      "reward_std": 0.1551719233393669,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.04296875,
+      "step": 47
+    },
+    {
+      "completion_length": 632.2890625,
+      "epoch": 0.05573294629898404,
+      "grad_norm": 1.2667105724676624,
+      "kl": 0.029102325439453125,
+      "learning_rate": 9.982881559634946e-07,
+      "loss": 0.0012,
+      "reward": 0.08203125,
+      "reward_std": 0.14755274169147015,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.05078125,
+      "step": 48
+    },
+    {
+      "completion_length": 571.69921875,
+      "epoch": 0.056894049346879534,
+      "grad_norm": 51.00112062221978,
+      "kl": 0.9501419067382812,
+      "learning_rate": 9.981290962387997e-07,
+      "loss": 0.0379,
+      "reward": 0.07421875,
+      "reward_std": 0.12532384134829044,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.05859375,
+      "step": 49
+    },
+    {
+      "completion_length": 595.40234375,
+      "epoch": 0.05805515239477504,
+      "grad_norm": 1.04581231978298,
+      "kl": 0.01190185546875,
+      "learning_rate": 9.979629852303425e-07,
+      "loss": 0.0005,
+      "reward": 0.11328125,
+      "reward_std": 0.18154333159327507,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.08203125,
+      "step": 50
+    },
+    {
+      "completion_length": 561.47265625,
+      "epoch": 0.059216255442670535,
+      "grad_norm": 0.7109293653864751,
+      "kl": 0.0103302001953125,
+      "learning_rate": 9.977898252895134e-07,
+      "loss": 0.0004,
+      "reward": 0.13671875,
+      "reward_std": 0.2260015867650509,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.09765625,
+      "step": 51
+    },
+    {
+      "completion_length": 589.95703125,
+      "epoch": 0.06037735849056604,
+      "grad_norm": 2.9062465124154735,
+      "kl": 0.00753021240234375,
+      "learning_rate": 9.976096188674836e-07,
+      "loss": 0.0003,
+      "reward": 0.12890625,
+      "reward_std": 0.19242635369300842,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.10546875,
+      "step": 52
+    },
+    {
+      "completion_length": 537.0234375,
+      "epoch": 0.06153846153846154,
+      "grad_norm": 22.409980149466673,
+      "kl": 0.2718505859375,
+      "learning_rate": 9.974223685151718e-07,
+      "loss": 0.0109,
+      "reward": 0.12890625,
+      "reward_std": 0.1836371347308159,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.10546875,
+      "step": 53
+    },
+    {
+      "completion_length": 531.80859375,
+      "epoch": 0.06269956458635705,
+      "grad_norm": 2.8826743442454297,
+      "kl": 0.122955322265625,
+      "learning_rate": 9.972280768832067e-07,
+      "loss": 0.0049,
+      "reward": 0.220703125,
+      "reward_std": 0.3008614853024483,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.126953125,
+      "step": 54
+    },
+    {
+      "completion_length": 611.5,
+      "epoch": 0.06386066763425254,
+      "grad_norm": 2.352779269743852,
+      "kl": 0.02130889892578125,
+      "learning_rate": 9.970267467218902e-07,
+      "loss": 0.0009,
+      "reward": 0.232421875,
+      "reward_std": 0.2863368093967438,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.177734375,
+      "step": 55
+    },
+    {
+      "completion_length": 555.91796875,
+      "epoch": 0.06502177068214804,
+      "grad_norm": 0.9680034082980018,
+      "kl": 0.01513671875,
+      "learning_rate": 9.968183808811585e-07,
+      "loss": 0.0006,
+      "reward": 0.24609375,
+      "reward_std": 0.3278784677386284,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.17578125,
+      "step": 56
+    },
+    {
+      "completion_length": 561.3203125,
+      "epoch": 0.06618287373004354,
+      "grad_norm": 5.155662797346911,
+      "kl": 0.1178741455078125,
+      "learning_rate": 9.966029823105415e-07,
+      "loss": 0.0047,
+      "reward": 0.291015625,
+      "reward_std": 0.3391275480389595,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.197265625,
+      "step": 57
+    },
+    {
+      "completion_length": 574.9453125,
+      "epoch": 0.06734397677793905,
+      "grad_norm": 3.5240176122453306,
+      "kl": 0.0246734619140625,
+      "learning_rate": 9.96380554059121e-07,
+      "loss": 0.001,
+      "reward": 0.291015625,
+      "reward_std": 0.31832827627658844,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.212890625,
+      "step": 58
+    },
+    {
+      "completion_length": 583.93359375,
+      "epoch": 0.06850507982583454,
+      "grad_norm": 3.558262716521494,
+      "kl": 0.0449981689453125,
+      "learning_rate": 9.961510992754882e-07,
+      "loss": 0.0018,
+      "reward": 0.28515625,
+      "reward_std": 0.26169369369745255,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.21484375,
+      "step": 59
+    },
+    {
+      "completion_length": 498.65625,
+      "epoch": 0.06966618287373004,
+      "grad_norm": 2.811935806633887,
+      "kl": 0.0194091796875,
+      "learning_rate": 9.959146212076978e-07,
+      "loss": 0.0008,
+      "reward": 0.33984375,
+      "reward_std": 0.314444862306118,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.27734375,
+      "step": 60
+    },
+    {
+      "completion_length": 478.53125,
+      "epoch": 0.07082728592162554,
+      "grad_norm": 4.984861468141312,
+      "kl": 0.107757568359375,
+      "learning_rate": 9.95671123203224e-07,
+      "loss": 0.0043,
+      "reward": 0.361328125,
+      "reward_std": 0.33764002099633217,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.267578125,
+      "step": 61
+    },
+    {
+      "completion_length": 512.39453125,
+      "epoch": 0.07198838896952105,
+      "grad_norm": 1.007778929339388,
+      "kl": 0.01971435546875,
+      "learning_rate": 9.954206087089105e-07,
+      "loss": 0.0008,
+      "reward": 0.3203125,
+      "reward_std": 0.2790137939155102,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.2734375,
+      "step": 62
+    },
+    {
+      "completion_length": 516.03515625,
+      "epoch": 0.07314949201741655,
+      "grad_norm": 0.5960945531101304,
+      "kl": 0.0254974365234375,
+      "learning_rate": 9.951630812709244e-07,
+      "loss": 0.001,
+      "reward": 0.373046875,
+      "reward_std": 0.3191210627555847,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.279296875,
+      "step": 63
+    },
+    {
+      "completion_length": 478.65625,
+      "epoch": 0.07431059506531204,
+      "grad_norm": 3.3120058314917094,
+      "kl": 0.027435302734375,
+      "learning_rate": 9.948985445347044e-07,
+      "loss": 0.0011,
+      "reward": 0.396484375,
+      "reward_std": 0.30096181109547615,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.326171875,
+      "step": 64
+    },
+    {
+      "completion_length": 453.5390625,
+      "epoch": 0.07547169811320754,
+      "grad_norm": 10.565800054920043,
+      "kl": 0.06463623046875,
+      "learning_rate": 9.946270022449093e-07,
+      "loss": 0.0026,
+      "reward": 0.40234375,
+      "reward_std": 0.30288316309452057,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.33203125,
+      "step": 65
+    },
+    {
+      "completion_length": 467.4609375,
+      "epoch": 0.07663280116110305,
+      "grad_norm": 8.848702328863897,
+      "kl": 0.026763916015625,
+      "learning_rate": 9.94348458245365e-07,
+      "loss": 0.0011,
+      "reward": 0.484375,
+      "reward_std": 0.3539407253265381,
+      "rewards/correctness_reward_func": 0.1640625,
+      "rewards/strict_format_reward_func": 0.3203125,
+      "step": 66
+    },
+    {
+      "completion_length": 460.51171875,
+      "epoch": 0.07779390420899855,
+      "grad_norm": 0.8276997968349737,
+      "kl": 0.028564453125,
+      "learning_rate": 9.940629164790118e-07,
+      "loss": 0.0011,
+      "reward": 0.390625,
+      "reward_std": 0.2685689851641655,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.3359375,
+      "step": 67
+    },
+    {
+      "completion_length": 432.30078125,
+      "epoch": 0.07895500725689406,
+      "grad_norm": 1.6501741381150759,
+      "kl": 0.068115234375,
+      "learning_rate": 9.937703809878454e-07,
+      "loss": 0.0027,
+      "reward": 0.447265625,
+      "reward_std": 0.32472314685583115,
+      "rewards/correctness_reward_func": 0.109375,
+      "rewards/strict_format_reward_func": 0.337890625,
+      "step": 68
+    },
+    {
+      "completion_length": 433.5625,
+      "epoch": 0.08011611030478955,
+      "grad_norm": 5.548040313902584,
+      "kl": 0.0538330078125,
+      "learning_rate": 9.934708559128622e-07,
+      "loss": 0.0022,
+      "reward": 0.48046875,
+      "reward_std": 0.3404357209801674,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.36328125,
+      "step": 69
+    },
+    {
+      "completion_length": 411.9921875,
+      "epoch": 0.08127721335268505,
+      "grad_norm": 4.54760002369526,
+      "kl": 0.0399169921875,
+      "learning_rate": 9.931643454939998e-07,
+      "loss": 0.0016,
+      "reward": 0.505859375,
+      "reward_std": 0.3411124534904957,
+      "rewards/correctness_reward_func": 0.1484375,
+      "rewards/strict_format_reward_func": 0.357421875,
+      "step": 70
+    },
+    {
+      "completion_length": 374.98828125,
+      "epoch": 0.08243831640058055,
+      "grad_norm": 0.7476462962662087,
+      "kl": 0.06536865234375,
+      "learning_rate": 9.928508540700772e-07,
+      "loss": 0.0026,
+      "reward": 0.458984375,
+      "reward_std": 0.2434551902115345,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.365234375,
+      "step": 71
+    },
+    {
+      "completion_length": 371.80859375,
+      "epoch": 0.08359941944847606,
+      "grad_norm": 0.31697483250235303,
+      "kl": 0.04022216796875,
+      "learning_rate": 9.925303860787333e-07,
+      "loss": 0.0016,
+      "reward": 0.427734375,
+      "reward_std": 0.24097520112991333,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.373046875,
+      "step": 72
+    },
+    {
+      "completion_length": 408.953125,
+      "epoch": 0.08476052249637156,
+      "grad_norm": 2.342541131973969,
+      "kl": 0.0394287109375,
+      "learning_rate": 9.92202946056364e-07,
+      "loss": 0.0016,
+      "reward": 0.447265625,
+      "reward_std": 0.2658774182200432,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.384765625,
+      "step": 73
+    },
+    {
+      "completion_length": 367.03125,
+      "epoch": 0.08592162554426705,
+      "grad_norm": 0.457004333832563,
+      "kl": 0.05194091796875,
+      "learning_rate": 9.918685386380572e-07,
+      "loss": 0.0021,
+      "reward": 0.478515625,
+      "reward_std": 0.22598736733198166,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.400390625,
+      "step": 74
+    },
+    {
+      "completion_length": 423.35546875,
+      "epoch": 0.08708272859216255,
+      "grad_norm": 81.83417853597544,
+      "kl": 0.247314453125,
+      "learning_rate": 9.915271685575296e-07,
+      "loss": 0.0099,
+      "reward": 0.48828125,
+      "reward_std": 0.3054451234638691,
+      "rewards/correctness_reward_func": 0.1015625,
+      "rewards/strict_format_reward_func": 0.38671875,
+      "step": 75
+    },
+    {
+      "completion_length": 312.484375,
+      "epoch": 0.08824383164005806,
+      "grad_norm": 1.780578379981416,
+      "kl": 0.0880126953125,
+      "learning_rate": 9.911788406470568e-07,
+      "loss": 0.0035,
+      "reward": 0.548828125,
+      "reward_std": 0.32934778556227684,
+      "rewards/correctness_reward_func": 0.125,
+      "rewards/strict_format_reward_func": 0.423828125,
+      "step": 76
+    },
+    {
+      "completion_length": 338.2421875,
+      "epoch": 0.08940493468795356,
+      "grad_norm": 0.6544200585420813,
+      "kl": 0.0731201171875,
+      "learning_rate": 9.908235598374068e-07,
+      "loss": 0.0029,
+      "reward": 0.45703125,
+      "reward_std": 0.1958453767001629,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.41796875,
+      "step": 77
+    },
+    {
+      "completion_length": 268.171875,
+      "epoch": 0.09056603773584905,
+      "grad_norm": 0.5879416696261314,
+      "kl": 0.108642578125,
+      "learning_rate": 9.904613311577695e-07,
+      "loss": 0.0044,
+      "reward": 0.556640625,
+      "reward_std": 0.22046153992414474,
+      "rewards/correctness_reward_func": 0.140625,
+      "rewards/strict_format_reward_func": 0.416015625,
+      "step": 78
+    },
+    {
+      "completion_length": 231.109375,
+      "epoch": 0.09172714078374455,
+      "grad_norm": 0.8976017470783989,
+      "kl": 0.18408203125,
+      "learning_rate": 9.900921597356855e-07,
+      "loss": 0.0074,
+      "reward": 0.515625,
+      "reward_std": 0.20618988201022148,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.4296875,
+      "step": 79
+    },
+    {
+      "completion_length": 196.59765625,
+      "epoch": 0.09288824383164006,
+      "grad_norm": 17.77907159035059,
+      "kl": 0.263916015625,
+      "learning_rate": 9.897160507969735e-07,
+      "loss": 0.0106,
+      "reward": 0.47265625,
+      "reward_std": 0.19209052622318268,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.41796875,
+      "step": 80
+    },
+    {
+      "completion_length": 110.4609375,
+      "epoch": 0.09404934687953556,
+      "grad_norm": 1.3276411102732406,
+      "kl": 0.3662109375,
+      "learning_rate": 9.893330096656573e-07,
+      "loss": 0.0147,
+      "reward": 0.578125,
+      "reward_std": 0.27833693847060204,
+      "rewards/correctness_reward_func": 0.140625,
+      "rewards/strict_format_reward_func": 0.4375,
+      "step": 81
+    },
+    {
+      "completion_length": 77.88671875,
+      "epoch": 0.09521044992743107,
+      "grad_norm": 1.9379785150930775,
+      "kl": 0.51953125,
+      "learning_rate": 9.889430417638883e-07,
+      "loss": 0.0208,
+      "reward": 0.51171875,
+      "reward_std": 0.1544804871082306,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.45703125,
+      "step": 82
+    },
+    {
+      "completion_length": 73.9375,
+      "epoch": 0.09637155297532655,
+      "grad_norm": 0.8473473359273695,
+      "kl": 0.48486328125,
+      "learning_rate": 9.885461526118713e-07,
+      "loss": 0.0194,
+      "reward": 0.50390625,
+      "reward_std": 0.14182795584201813,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.45703125,
+      "step": 83
+    },
+    {
+      "completion_length": 56.2109375,
+      "epoch": 0.09753265602322206,
+      "grad_norm": 2.281085454498393,
+      "kl": 0.49072265625,
+      "learning_rate": 9.88142347827784e-07,
+      "loss": 0.0196,
+      "reward": 0.4765625,
+      "reward_std": 0.12851098738610744,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.453125,
+      "step": 84
+    },
+    {
+      "completion_length": 37.875,
+      "epoch": 0.09869375907111756,
+      "grad_norm": 0.5788019582854786,
+      "kl": 0.5927734375,
+      "learning_rate": 9.877316331276993e-07,
+      "loss": 0.0237,
+      "reward": 0.470703125,
+      "reward_std": 0.055291797965765,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.470703125,
+      "step": 85
+    },
+    {
+      "completion_length": 33.50390625,
+      "epoch": 0.09985486211901307,
+      "grad_norm": 6.104259815482374,
+      "kl": 0.5439453125,
+      "learning_rate": 9.873140143255034e-07,
+      "loss": 0.0217,
+      "reward": 0.517578125,
+      "reward_std": 0.07212539482861757,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.470703125,
+      "step": 86
+    },
+    {
+      "completion_length": 20.76171875,
+      "epoch": 0.10101596516690857,
+      "grad_norm": 0.507931311814518,
+      "kl": 0.5947265625,
+      "learning_rate": 9.86889497332814e-07,
+      "loss": 0.0238,
+      "reward": 0.486328125,
+      "reward_std": 0.02734375,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 87
+    },
+    {
+      "completion_length": 16.58203125,
+      "epoch": 0.10217706821480406,
+      "grad_norm": 0.45325431756004353,
+      "kl": 0.6630859375,
+      "learning_rate": 9.864580881588958e-07,
+      "loss": 0.0265,
+      "reward": 0.501953125,
+      "reward_std": 0.024041798897087574,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.494140625,
+      "step": 88
+    },
+    {
+      "completion_length": 16.12109375,
+      "epoch": 0.10333817126269956,
+      "grad_norm": 0.38014072113379394,
+      "kl": 0.6513671875,
+      "learning_rate": 9.860197929105767e-07,
+      "loss": 0.0261,
+      "reward": 0.49609375,
+      "reward_std": 0.0078125,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.49609375,
+      "step": 89
+    },
+    {
+      "completion_length": 19.7265625,
+      "epoch": 0.10449927431059507,
+      "grad_norm": 3.246980620297076,
+      "kl": 0.6064453125,
+      "learning_rate": 9.8557461779216e-07,
+      "loss": 0.0243,
+      "reward": 0.48828125,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 90
+    },
+    {
+      "completion_length": 15.8359375,
+      "epoch": 0.10566037735849057,
+      "grad_norm": 1.125767741710113,
+      "kl": 0.6162109375,
+      "learning_rate": 9.85122569105338e-07,
+      "loss": 0.0247,
+      "reward": 0.49609375,
+      "reward_std": 0.0078125,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.49609375,
+      "step": 91
+    },
+    {
+      "completion_length": 22.44921875,
+      "epoch": 0.10682148040638606,
+      "grad_norm": 0.44874987843781083,
+      "kl": 0.6083984375,
+      "learning_rate": 9.846636532491012e-07,
+      "loss": 0.0243,
+      "reward": 0.48828125,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 92
+    },
+    {
+      "completion_length": 16.21484375,
+      "epoch": 0.10798258345428156,
+      "grad_norm": 2.7244778647827363,
+      "kl": 0.796875,
+      "learning_rate": 9.841978767196493e-07,
+      "loss": 0.0319,
+      "reward": 0.494140625,
+      "reward_std": 0.01171875,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.494140625,
+      "step": 93
+    },
+    {
+      "completion_length": 18.08984375,
+      "epoch": 0.10914368650217707,
+      "grad_norm": 0.900717061581606,
+      "kl": 0.5712890625,
+      "learning_rate": 9.83725246110298e-07,
+      "loss": 0.0229,
+      "reward": 0.49609375,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 94
+    },
+    {
+      "completion_length": 19.921875,
+      "epoch": 0.11030478955007257,
+      "grad_norm": 0.2922094758318377,
+      "kl": 0.62109375,
+      "learning_rate": 9.832457681113865e-07,
+      "loss": 0.0249,
+      "reward": 0.51953125,
+      "reward_std": 0.0546875,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 95
+    },
+    {
+      "completion_length": 22.48828125,
+      "epoch": 0.11146589259796807,
+      "grad_norm": 1.5719055280514225,
+      "kl": 0.5966796875,
+      "learning_rate": 9.827594495101822e-07,
+      "loss": 0.0239,
+      "reward": 0.53125,
+      "reward_std": 0.049292195588350296,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.4921875,
+      "step": 96
+    },
+    {
+      "completion_length": 21.078125,
+      "epoch": 0.11262699564586356,
+      "grad_norm": 1.875138112811585,
+      "kl": 0.62890625,
+      "learning_rate": 9.822662971907852e-07,
+      "loss": 0.0252,
+      "reward": 0.505859375,
+      "reward_std": 0.047479298897087574,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 97
+    },
+    {
+      "completion_length": 20.328125,
+      "epoch": 0.11378809869375907,
+      "grad_norm": 0.46440948422543626,
+      "kl": 0.5703125,
+      "learning_rate": 9.8176631813403e-07,
+      "loss": 0.0228,
+      "reward": 0.5078125,
+      "reward_std": 0.015625,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.5,
+      "step": 98
+    },
+    {
+      "completion_length": 17.6953125,
+      "epoch": 0.11494920174165457,
+      "grad_norm": 2.350666772323524,
+      "kl": 0.6298828125,
+      "learning_rate": 9.812595194173874e-07,
+      "loss": 0.0252,
+      "reward": 0.50390625,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.49609375,
+      "step": 99
+    },
+    {
+      "completion_length": 30.88671875,
+      "epoch": 0.11611030478955008,
+      "grad_norm": 0.6769899791454503,
+      "kl": 0.537109375,
+      "learning_rate": 9.807459082148648e-07,
+      "loss": 0.0215,
+      "reward": 0.56640625,
+      "reward_std": 0.0804273895919323,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 100
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 861,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3ef0b57332a5361e40cb4439506d8b573d5a0adf1890f8b03706312320d537d
+size 6904

checkpoint-200/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-200/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-1.5B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.3",
+  "use_cache": true,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

checkpoint-200/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.48.3"
+}

checkpoint-200/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51d9a9e14322f1dfaa9ff9d3f3dd857e34c7f3a9f74e435d3e11202bfdeab11a
+size 3087467144

checkpoint-200/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>"
+}

checkpoint-200/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

checkpoint-200/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-200/trainer_state.json

@@ -0,0 +1,2633 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.23222060957910015,
+  "eval_steps": 500,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 724.58984375,
+      "epoch": 0.0011611030478955006,
+      "grad_norm": 0.7112110585443041,
+      "kl": 0.0009012222290039062,
+      "learning_rate": 3.846153846153846e-08,
+      "loss": 0.0,
+      "reward": 0.046875,
+      "reward_std": 0.06733439117670059,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 1
+    },
+    {
+      "completion_length": 708.56640625,
+      "epoch": 0.0023222060957910013,
+      "grad_norm": 0.12068406687948798,
+      "kl": 0.000453948974609375,
+      "learning_rate": 7.692307692307692e-08,
+      "loss": 0.0,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 2
+    },
+    {
+      "completion_length": 742.65625,
+      "epoch": 0.0034833091436865023,
+      "grad_norm": 0.7976659386050398,
+      "kl": 0.0007343292236328125,
+      "learning_rate": 1.1538461538461539e-07,
+      "loss": 0.0,
+      "reward": 0.0234375,
+      "reward_std": 0.046875,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 3
+    },
+    {
+      "completion_length": 808.21875,
+      "epoch": 0.0046444121915820025,
+      "grad_norm": 0.40112277888275455,
+      "kl": 0.00063323974609375,
+      "learning_rate": 1.5384615384615385e-07,
+      "loss": 0.0,
+      "reward": 0.060546875,
+      "reward_std": 0.1078859455883503,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 4
+    },
+    {
+      "completion_length": 654.234375,
+      "epoch": 0.005805515239477504,
+      "grad_norm": 0.39725912723012635,
+      "kl": 0.0004980564117431641,
+      "learning_rate": 1.9230769230769231e-07,
+      "loss": 0.0,
+      "reward": 0.044921875,
+      "reward_std": 0.08984375,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 5
+    },
+    {
+      "completion_length": 772.68359375,
+      "epoch": 0.006966618287373005,
+      "grad_norm": 0.14294736924830026,
+      "kl": 0.0004153251647949219,
+      "learning_rate": 2.3076923076923078e-07,
+      "loss": 0.0,
+      "reward": 0.046875,
+      "reward_std": 0.0805421955883503,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 6
+    },
+    {
+      "completion_length": 745.203125,
+      "epoch": 0.008127721335268505,
+      "grad_norm": 0.16815060965909404,
+      "kl": 0.0009188652038574219,
+      "learning_rate": 2.692307692307692e-07,
+      "loss": 0.0,
+      "reward": 0.021484375,
+      "reward_std": 0.029760945588350296,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 7
+    },
+    {
+      "completion_length": 744.265625,
+      "epoch": 0.009288824383164005,
+      "grad_norm": 12.779442066715234,
+      "kl": 0.00357818603515625,
+      "learning_rate": 3.076923076923077e-07,
+      "loss": 0.0001,
+      "reward": 0.03125,
+      "reward_std": 0.049292195588350296,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.0,
+      "step": 8
+    },
+    {
+      "completion_length": 704.51953125,
+      "epoch": 0.010449927431059507,
+      "grad_norm": 0.2475578183097461,
+      "kl": 0.0035419464111328125,
+      "learning_rate": 3.461538461538461e-07,
+      "loss": 0.0001,
+      "reward": 0.060546875,
+      "reward_std": 0.1078859455883503,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 9
+    },
+    {
+      "completion_length": 759.2890625,
+      "epoch": 0.011611030478955007,
+      "grad_norm": 0.3357083329970443,
+      "kl": 0.00310516357421875,
+      "learning_rate": 3.8461538461538463e-07,
+      "loss": 0.0001,
+      "reward": 0.00390625,
+      "reward_std": 0.0078125,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.00390625,
+      "step": 10
+    },
+    {
+      "completion_length": 746.0703125,
+      "epoch": 0.012772133526850507,
+      "grad_norm": 0.2657165785401948,
+      "kl": 0.00334930419921875,
+      "learning_rate": 4.2307692307692304e-07,
+      "loss": 0.0001,
+      "reward": 0.017578125,
+      "reward_std": 0.03515625,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 11
+    },
+    {
+      "completion_length": 702.21875,
+      "epoch": 0.01393323657474601,
+      "grad_norm": 1.2656634353615759,
+      "kl": 0.0019159317016601562,
+      "learning_rate": 4.6153846153846156e-07,
+      "loss": 0.0001,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 12
+    },
+    {
+      "completion_length": 735.78515625,
+      "epoch": 0.01509433962264151,
+      "grad_norm": 0.5357650576255997,
+      "kl": 0.005596160888671875,
+      "learning_rate": 5e-07,
+      "loss": 0.0002,
+      "reward": 0.06640625,
+      "reward_std": 0.1280700732022524,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.01171875,
+      "step": 13
+    },
+    {
+      "completion_length": 802.51953125,
+      "epoch": 0.01625544267053701,
+      "grad_norm": 1.9685839556774372,
+      "kl": 0.00266265869140625,
+      "learning_rate": 5.384615384615384e-07,
+      "loss": 0.0001,
+      "reward": 0.03125,
+      "reward_std": 0.059198048897087574,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.015625,
+      "step": 14
+    },
+    {
+      "completion_length": 702.47265625,
+      "epoch": 0.01741654571843251,
+      "grad_norm": 0.3022926592468717,
+      "kl": 0.00206756591796875,
+      "learning_rate": 5.769230769230768e-07,
+      "loss": 0.0001,
+      "reward": 0.03125,
+      "reward_std": 0.049292195588350296,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 15
+    },
+    {
+      "completion_length": 714.55859375,
+      "epoch": 0.01857764876632801,
+      "grad_norm": 0.67295700611606,
+      "kl": 0.006526947021484375,
+      "learning_rate": 6.153846153846154e-07,
+      "loss": 0.0003,
+      "reward": 0.0546875,
+      "reward_std": 0.04213257320225239,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 16
+    },
+    {
+      "completion_length": 728.94921875,
+      "epoch": 0.019738751814223514,
+      "grad_norm": 0.12683035812623522,
+      "kl": 0.003253936767578125,
+      "learning_rate": 6.538461538461538e-07,
+      "loss": 0.0001,
+      "reward": 0.041015625,
+      "reward_std": 0.0688234455883503,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 17
+    },
+    {
+      "completion_length": 752.4609375,
+      "epoch": 0.020899854862119014,
+      "grad_norm": 0.7572918382078203,
+      "kl": 0.002063751220703125,
+      "learning_rate": 6.923076923076922e-07,
+      "loss": 0.0001,
+      "reward": 0.01953125,
+      "reward_std": 0.0390625,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.00390625,
+      "step": 18
+    },
+    {
+      "completion_length": 686.39453125,
+      "epoch": 0.022060957910014514,
+      "grad_norm": 0.366311042586646,
+      "kl": 0.00254058837890625,
+      "learning_rate": 7.307692307692307e-07,
+      "loss": 0.0001,
+      "reward": 0.037109375,
+      "reward_std": 0.07421875,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 19
+    },
+    {
+      "completion_length": 814.58203125,
+      "epoch": 0.023222060957910014,
+      "grad_norm": 0.39037265038987773,
+      "kl": 0.0024509429931640625,
+      "learning_rate": 7.692307692307693e-07,
+      "loss": 0.0001,
+      "reward": 0.056640625,
+      "reward_std": 0.1000734455883503,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 20
+    },
+    {
+      "completion_length": 790.58203125,
+      "epoch": 0.024383164005805515,
+      "grad_norm": 0.16449028843577287,
+      "kl": 0.0017910003662109375,
+      "learning_rate": 8.076923076923077e-07,
+      "loss": 0.0001,
+      "reward": 0.021484375,
+      "reward_std": 0.04296875,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.005859375,
+      "step": 21
+    },
+    {
+      "completion_length": 720.109375,
+      "epoch": 0.025544267053701015,
+      "grad_norm": 0.16050190810020273,
+      "kl": 0.000865936279296875,
+      "learning_rate": 8.461538461538461e-07,
+      "loss": 0.0,
+      "reward": 0.041015625,
+      "reward_std": 0.08203125,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 22
+    },
+    {
+      "completion_length": 772.921875,
+      "epoch": 0.026705370101596515,
+      "grad_norm": 1.022693227541973,
+      "kl": 0.010603904724121094,
+      "learning_rate": 8.846153846153846e-07,
+      "loss": 0.0004,
+      "reward": 0.0234375,
+      "reward_std": 0.046875,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 23
+    },
+    {
+      "completion_length": 785.9296875,
+      "epoch": 0.02786647314949202,
+      "grad_norm": 0.4975098781878154,
+      "kl": 0.0006268024444580078,
+      "learning_rate": 9.230769230769231e-07,
+      "loss": 0.0,
+      "reward": 0.0546875,
+      "reward_std": 0.10463257133960724,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.0078125,
+      "step": 24
+    },
+    {
+      "completion_length": 808.91796875,
+      "epoch": 0.02902757619738752,
+      "grad_norm": 0.07636536312105723,
+      "kl": 0.0006952285766601562,
+      "learning_rate": 9.615384615384615e-07,
+      "loss": 0.0,
+      "reward": 0.03515625,
+      "reward_std": 0.055702777579426765,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.01171875,
+      "step": 25
+    },
+    {
+      "completion_length": 677.65234375,
+      "epoch": 0.03018867924528302,
+      "grad_norm": 105.92855708885655,
+      "kl": 0.048847198486328125,
+      "learning_rate": 1e-06,
+      "loss": 0.002,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.017578125,
+      "step": 26
+    },
+    {
+      "completion_length": 657.85546875,
+      "epoch": 0.03134978229317852,
+      "grad_norm": 0.0861458003290139,
+      "kl": 0.004261970520019531,
+      "learning_rate": 9.999964611162971e-07,
+      "loss": 0.0002,
+      "reward": 0.009765625,
+      "reward_std": 0.01953125,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 27
+    },
+    {
+      "completion_length": 726.2421875,
+      "epoch": 0.03251088534107402,
+      "grad_norm": 0.14527394620380304,
+      "kl": 0.00072479248046875,
+      "learning_rate": 9.999858445152839e-07,
+      "loss": 0.0,
+      "reward": 0.033203125,
+      "reward_std": 0.053198445588350296,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.001953125,
+      "step": 28
+    },
+    {
+      "completion_length": 646.53125,
+      "epoch": 0.03367198838896952,
+      "grad_norm": 0.4038815325919152,
+      "kl": 0.008038520812988281,
+      "learning_rate": 9.999681503472433e-07,
+      "loss": 0.0003,
+      "reward": 0.02734375,
+      "reward_std": 0.043573048897087574,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.01171875,
+      "step": 29
+    },
+    {
+      "completion_length": 676.234375,
+      "epoch": 0.03483309143686502,
+      "grad_norm": 24.530460542475996,
+      "kl": 0.08559513092041016,
+      "learning_rate": 9.99943378862646e-07,
+      "loss": 0.0034,
+      "reward": 0.076171875,
+      "reward_std": 0.1358339935541153,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 30
+    },
+    {
+      "completion_length": 669.62890625,
+      "epoch": 0.035994194484760524,
+      "grad_norm": 12.948432597412872,
+      "kl": 0.0016450881958007812,
+      "learning_rate": 9.999115304121457e-07,
+      "loss": 0.0001,
+      "reward": 0.033203125,
+      "reward_std": 0.06640625,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.009765625,
+      "step": 31
+    },
+    {
+      "completion_length": 632.80078125,
+      "epoch": 0.03715529753265602,
+      "grad_norm": 0.14738843577567481,
+      "kl": 0.0011644363403320312,
+      "learning_rate": 9.998726054465744e-07,
+      "loss": 0.0,
+      "reward": 0.0546875,
+      "reward_std": 0.106073047965765,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.015625,
+      "step": 32
+    },
+    {
+      "completion_length": 686.90234375,
+      "epoch": 0.038316400580551524,
+      "grad_norm": 6.995419155886516,
+      "kl": 0.056095123291015625,
+      "learning_rate": 9.998266045169354e-07,
+      "loss": 0.0022,
+      "reward": 0.052734375,
+      "reward_std": 0.0922609455883503,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 33
+    },
+    {
+      "completion_length": 655.90234375,
+      "epoch": 0.03947750362844703,
+      "grad_norm": 3.5735767009613224,
+      "kl": 0.013265609741210938,
+      "learning_rate": 9.997735282743968e-07,
+      "loss": 0.0005,
+      "reward": 0.076171875,
+      "reward_std": 0.13439351692795753,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.013671875,
+      "step": 34
+    },
+    {
+      "completion_length": 639.82421875,
+      "epoch": 0.040638606676342524,
+      "grad_norm": 1.8475352137430245,
+      "kl": 0.0025959014892578125,
+      "learning_rate": 9.997133774702812e-07,
+      "loss": 0.0001,
+      "reward": 0.05078125,
+      "reward_std": 0.09682007133960724,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.01953125,
+      "step": 35
+    },
+    {
+      "completion_length": 715.3125,
+      "epoch": 0.04179970972423803,
+      "grad_norm": 13.632018237363075,
+      "kl": 0.21777725219726562,
+      "learning_rate": 9.996461529560552e-07,
+      "loss": 0.0087,
+      "reward": 0.0390625,
+      "reward_std": 0.078125,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.015625,
+      "step": 36
+    },
+    {
+      "completion_length": 631.8203125,
+      "epoch": 0.042960812772133525,
+      "grad_norm": 10.015816840984483,
+      "kl": 0.00942230224609375,
+      "learning_rate": 9.995718556833178e-07,
+      "loss": 0.0004,
+      "reward": 0.072265625,
+      "reward_std": 0.141229297965765,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.025390625,
+      "step": 37
+    },
+    {
+      "completion_length": 636.9140625,
+      "epoch": 0.04412191582002903,
+      "grad_norm": 0.18443450708075712,
+      "kl": 0.001720428466796875,
+      "learning_rate": 9.994904867037865e-07,
+      "loss": 0.0001,
+      "reward": 0.033203125,
+      "reward_std": 0.046594543382525444,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.017578125,
+      "step": 38
+    },
+    {
+      "completion_length": 667.6484375,
+      "epoch": 0.045283018867924525,
+      "grad_norm": 0.7286107675191874,
+      "kl": 0.027004241943359375,
+      "learning_rate": 9.994020471692832e-07,
+      "loss": 0.0011,
+      "reward": 0.0625,
+      "reward_std": 0.10044586285948753,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.0234375,
+      "step": 39
+    },
+    {
+      "completion_length": 695.68359375,
+      "epoch": 0.04644412191582003,
+      "grad_norm": 0.7697487269128109,
+      "kl": 0.002681732177734375,
+      "learning_rate": 9.993065383317162e-07,
+      "loss": 0.0001,
+      "reward": 0.078125,
+      "reward_std": 0.12407248839735985,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.0234375,
+      "step": 40
+    },
+    {
+      "completion_length": 678.390625,
+      "epoch": 0.04760522496371553,
+      "grad_norm": 0.33385332935798095,
+      "kl": 0.0028896331787109375,
+      "learning_rate": 9.992039615430648e-07,
+      "loss": 0.0001,
+      "reward": 0.06640625,
+      "reward_std": 0.1137621309608221,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.01953125,
+      "step": 41
+    },
+    {
+      "completion_length": 660.2421875,
+      "epoch": 0.04876632801161103,
+      "grad_norm": 5.2234598338711615,
+      "kl": 0.03473663330078125,
+      "learning_rate": 9.990943182553578e-07,
+      "loss": 0.0014,
+      "reward": 0.05859375,
+      "reward_std": 0.11058359593153,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.03515625,
+      "step": 42
+    },
+    {
+      "completion_length": 599.66015625,
+      "epoch": 0.04992743105950653,
+      "grad_norm": 0.3125631745813204,
+      "kl": 0.0029144287109375,
+      "learning_rate": 9.989776100206547e-07,
+      "loss": 0.0001,
+      "reward": 0.087890625,
+      "reward_std": 0.1592714935541153,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.033203125,
+      "step": 43
+    },
+    {
+      "completion_length": 695.41015625,
+      "epoch": 0.05108853410740203,
+      "grad_norm": 0.31129440427202076,
+      "kl": 0.002468109130859375,
+      "learning_rate": 9.98853838491023e-07,
+      "loss": 0.0001,
+      "reward": 0.046875,
+      "reward_std": 0.08240367192775011,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.0234375,
+      "step": 44
+    },
+    {
+      "completion_length": 594.67578125,
+      "epoch": 0.05224963715529753,
+      "grad_norm": 0.2962376329942141,
+      "kl": 0.00382232666015625,
+      "learning_rate": 9.98723005418515e-07,
+      "loss": 0.0002,
+      "reward": 0.091796875,
+      "reward_std": 0.1253724191337824,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.044921875,
+      "step": 45
+    },
+    {
+      "completion_length": 672.73046875,
+      "epoch": 0.05341074020319303,
+      "grad_norm": 3.778170631942383,
+      "kl": 0.077880859375,
+      "learning_rate": 9.985851126551428e-07,
+      "loss": 0.0031,
+      "reward": 0.103515625,
+      "reward_std": 0.1781556848436594,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.041015625,
+      "step": 46
+    },
+    {
+      "completion_length": 603.15625,
+      "epoch": 0.054571843251088534,
+      "grad_norm": 0.3069672598383264,
+      "kl": 0.006290435791015625,
+      "learning_rate": 9.98440162152852e-07,
+      "loss": 0.0003,
+      "reward": 0.08984375,
+      "reward_std": 0.1551719233393669,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.04296875,
+      "step": 47
+    },
+    {
+      "completion_length": 632.2890625,
+      "epoch": 0.05573294629898404,
+      "grad_norm": 1.2667105724676624,
+      "kl": 0.029102325439453125,
+      "learning_rate": 9.982881559634946e-07,
+      "loss": 0.0012,
+      "reward": 0.08203125,
+      "reward_std": 0.14755274169147015,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.05078125,
+      "step": 48
+    },
+    {
+      "completion_length": 571.69921875,
+      "epoch": 0.056894049346879534,
+      "grad_norm": 51.00112062221978,
+      "kl": 0.9501419067382812,
+      "learning_rate": 9.981290962387997e-07,
+      "loss": 0.0379,
+      "reward": 0.07421875,
+      "reward_std": 0.12532384134829044,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.05859375,
+      "step": 49
+    },
+    {
+      "completion_length": 595.40234375,
+      "epoch": 0.05805515239477504,
+      "grad_norm": 1.04581231978298,
+      "kl": 0.01190185546875,
+      "learning_rate": 9.979629852303425e-07,
+      "loss": 0.0005,
+      "reward": 0.11328125,
+      "reward_std": 0.18154333159327507,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.08203125,
+      "step": 50
+    },
+    {
+      "completion_length": 561.47265625,
+      "epoch": 0.059216255442670535,
+      "grad_norm": 0.7109293653864751,
+      "kl": 0.0103302001953125,
+      "learning_rate": 9.977898252895134e-07,
+      "loss": 0.0004,
+      "reward": 0.13671875,
+      "reward_std": 0.2260015867650509,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.09765625,
+      "step": 51
+    },
+    {
+      "completion_length": 589.95703125,
+      "epoch": 0.06037735849056604,
+      "grad_norm": 2.9062465124154735,
+      "kl": 0.00753021240234375,
+      "learning_rate": 9.976096188674836e-07,
+      "loss": 0.0003,
+      "reward": 0.12890625,
+      "reward_std": 0.19242635369300842,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.10546875,
+      "step": 52
+    },
+    {
+      "completion_length": 537.0234375,
+      "epoch": 0.06153846153846154,
+      "grad_norm": 22.409980149466673,
+      "kl": 0.2718505859375,
+      "learning_rate": 9.974223685151718e-07,
+      "loss": 0.0109,
+      "reward": 0.12890625,
+      "reward_std": 0.1836371347308159,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.10546875,
+      "step": 53
+    },
+    {
+      "completion_length": 531.80859375,
+      "epoch": 0.06269956458635705,
+      "grad_norm": 2.8826743442454297,
+      "kl": 0.122955322265625,
+      "learning_rate": 9.972280768832067e-07,
+      "loss": 0.0049,
+      "reward": 0.220703125,
+      "reward_std": 0.3008614853024483,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.126953125,
+      "step": 54
+    },
+    {
+      "completion_length": 611.5,
+      "epoch": 0.06386066763425254,
+      "grad_norm": 2.352779269743852,
+      "kl": 0.02130889892578125,
+      "learning_rate": 9.970267467218902e-07,
+      "loss": 0.0009,
+      "reward": 0.232421875,
+      "reward_std": 0.2863368093967438,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.177734375,
+      "step": 55
+    },
+    {
+      "completion_length": 555.91796875,
+      "epoch": 0.06502177068214804,
+      "grad_norm": 0.9680034082980018,
+      "kl": 0.01513671875,
+      "learning_rate": 9.968183808811585e-07,
+      "loss": 0.0006,
+      "reward": 0.24609375,
+      "reward_std": 0.3278784677386284,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.17578125,
+      "step": 56
+    },
+    {
+      "completion_length": 561.3203125,
+      "epoch": 0.06618287373004354,
+      "grad_norm": 5.155662797346911,
+      "kl": 0.1178741455078125,
+      "learning_rate": 9.966029823105415e-07,
+      "loss": 0.0047,
+      "reward": 0.291015625,
+      "reward_std": 0.3391275480389595,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.197265625,
+      "step": 57
+    },
+    {
+      "completion_length": 574.9453125,
+      "epoch": 0.06734397677793905,
+      "grad_norm": 3.5240176122453306,
+      "kl": 0.0246734619140625,
+      "learning_rate": 9.96380554059121e-07,
+      "loss": 0.001,
+      "reward": 0.291015625,
+      "reward_std": 0.31832827627658844,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.212890625,
+      "step": 58
+    },
+    {
+      "completion_length": 583.93359375,
+      "epoch": 0.06850507982583454,
+      "grad_norm": 3.558262716521494,
+      "kl": 0.0449981689453125,
+      "learning_rate": 9.961510992754882e-07,
+      "loss": 0.0018,
+      "reward": 0.28515625,
+      "reward_std": 0.26169369369745255,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.21484375,
+      "step": 59
+    },
+    {
+      "completion_length": 498.65625,
+      "epoch": 0.06966618287373004,
+      "grad_norm": 2.811935806633887,
+      "kl": 0.0194091796875,
+      "learning_rate": 9.959146212076978e-07,
+      "loss": 0.0008,
+      "reward": 0.33984375,
+      "reward_std": 0.314444862306118,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.27734375,
+      "step": 60
+    },
+    {
+      "completion_length": 478.53125,
+      "epoch": 0.07082728592162554,
+      "grad_norm": 4.984861468141312,
+      "kl": 0.107757568359375,
+      "learning_rate": 9.95671123203224e-07,
+      "loss": 0.0043,
+      "reward": 0.361328125,
+      "reward_std": 0.33764002099633217,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.267578125,
+      "step": 61
+    },
+    {
+      "completion_length": 512.39453125,
+      "epoch": 0.07198838896952105,
+      "grad_norm": 1.007778929339388,
+      "kl": 0.01971435546875,
+      "learning_rate": 9.954206087089105e-07,
+      "loss": 0.0008,
+      "reward": 0.3203125,
+      "reward_std": 0.2790137939155102,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.2734375,
+      "step": 62
+    },
+    {
+      "completion_length": 516.03515625,
+      "epoch": 0.07314949201741655,
+      "grad_norm": 0.5960945531101304,
+      "kl": 0.0254974365234375,
+      "learning_rate": 9.951630812709244e-07,
+      "loss": 0.001,
+      "reward": 0.373046875,
+      "reward_std": 0.3191210627555847,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.279296875,
+      "step": 63
+    },
+    {
+      "completion_length": 478.65625,
+      "epoch": 0.07431059506531204,
+      "grad_norm": 3.3120058314917094,
+      "kl": 0.027435302734375,
+      "learning_rate": 9.948985445347044e-07,
+      "loss": 0.0011,
+      "reward": 0.396484375,
+      "reward_std": 0.30096181109547615,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.326171875,
+      "step": 64
+    },
+    {
+      "completion_length": 453.5390625,
+      "epoch": 0.07547169811320754,
+      "grad_norm": 10.565800054920043,
+      "kl": 0.06463623046875,
+      "learning_rate": 9.946270022449093e-07,
+      "loss": 0.0026,
+      "reward": 0.40234375,
+      "reward_std": 0.30288316309452057,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.33203125,
+      "step": 65
+    },
+    {
+      "completion_length": 467.4609375,
+      "epoch": 0.07663280116110305,
+      "grad_norm": 8.848702328863897,
+      "kl": 0.026763916015625,
+      "learning_rate": 9.94348458245365e-07,
+      "loss": 0.0011,
+      "reward": 0.484375,
+      "reward_std": 0.3539407253265381,
+      "rewards/correctness_reward_func": 0.1640625,
+      "rewards/strict_format_reward_func": 0.3203125,
+      "step": 66
+    },
+    {
+      "completion_length": 460.51171875,
+      "epoch": 0.07779390420899855,
+      "grad_norm": 0.8276997968349737,
+      "kl": 0.028564453125,
+      "learning_rate": 9.940629164790118e-07,
+      "loss": 0.0011,
+      "reward": 0.390625,
+      "reward_std": 0.2685689851641655,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.3359375,
+      "step": 67
+    },
+    {
+      "completion_length": 432.30078125,
+      "epoch": 0.07895500725689406,
+      "grad_norm": 1.6501741381150759,
+      "kl": 0.068115234375,
+      "learning_rate": 9.937703809878454e-07,
+      "loss": 0.0027,
+      "reward": 0.447265625,
+      "reward_std": 0.32472314685583115,
+      "rewards/correctness_reward_func": 0.109375,
+      "rewards/strict_format_reward_func": 0.337890625,
+      "step": 68
+    },
+    {
+      "completion_length": 433.5625,
+      "epoch": 0.08011611030478955,
+      "grad_norm": 5.548040313902584,
+      "kl": 0.0538330078125,
+      "learning_rate": 9.934708559128622e-07,
+      "loss": 0.0022,
+      "reward": 0.48046875,
+      "reward_std": 0.3404357209801674,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.36328125,
+      "step": 69
+    },
+    {
+      "completion_length": 411.9921875,
+      "epoch": 0.08127721335268505,
+      "grad_norm": 4.54760002369526,
+      "kl": 0.0399169921875,
+      "learning_rate": 9.931643454939998e-07,
+      "loss": 0.0016,
+      "reward": 0.505859375,
+      "reward_std": 0.3411124534904957,
+      "rewards/correctness_reward_func": 0.1484375,
+      "rewards/strict_format_reward_func": 0.357421875,
+      "step": 70
+    },
+    {
+      "completion_length": 374.98828125,
+      "epoch": 0.08243831640058055,
+      "grad_norm": 0.7476462962662087,
+      "kl": 0.06536865234375,
+      "learning_rate": 9.928508540700772e-07,
+      "loss": 0.0026,
+      "reward": 0.458984375,
+      "reward_std": 0.2434551902115345,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.365234375,
+      "step": 71
+    },
+    {
+      "completion_length": 371.80859375,
+      "epoch": 0.08359941944847606,
+      "grad_norm": 0.31697483250235303,
+      "kl": 0.04022216796875,
+      "learning_rate": 9.925303860787333e-07,
+      "loss": 0.0016,
+      "reward": 0.427734375,
+      "reward_std": 0.24097520112991333,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.373046875,
+      "step": 72
+    },
+    {
+      "completion_length": 408.953125,
+      "epoch": 0.08476052249637156,
+      "grad_norm": 2.342541131973969,
+      "kl": 0.0394287109375,
+      "learning_rate": 9.92202946056364e-07,
+      "loss": 0.0016,
+      "reward": 0.447265625,
+      "reward_std": 0.2658774182200432,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.384765625,
+      "step": 73
+    },
+    {
+      "completion_length": 367.03125,
+      "epoch": 0.08592162554426705,
+      "grad_norm": 0.457004333832563,
+      "kl": 0.05194091796875,
+      "learning_rate": 9.918685386380572e-07,
+      "loss": 0.0021,
+      "reward": 0.478515625,
+      "reward_std": 0.22598736733198166,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.400390625,
+      "step": 74
+    },
+    {
+      "completion_length": 423.35546875,
+      "epoch": 0.08708272859216255,
+      "grad_norm": 81.83417853597544,
+      "kl": 0.247314453125,
+      "learning_rate": 9.915271685575296e-07,
+      "loss": 0.0099,
+      "reward": 0.48828125,
+      "reward_std": 0.3054451234638691,
+      "rewards/correctness_reward_func": 0.1015625,
+      "rewards/strict_format_reward_func": 0.38671875,
+      "step": 75
+    },
+    {
+      "completion_length": 312.484375,
+      "epoch": 0.08824383164005806,
+      "grad_norm": 1.780578379981416,
+      "kl": 0.0880126953125,
+      "learning_rate": 9.911788406470568e-07,
+      "loss": 0.0035,
+      "reward": 0.548828125,
+      "reward_std": 0.32934778556227684,
+      "rewards/correctness_reward_func": 0.125,
+      "rewards/strict_format_reward_func": 0.423828125,
+      "step": 76
+    },
+    {
+      "completion_length": 338.2421875,
+      "epoch": 0.08940493468795356,
+      "grad_norm": 0.6544200585420813,
+      "kl": 0.0731201171875,
+      "learning_rate": 9.908235598374068e-07,
+      "loss": 0.0029,
+      "reward": 0.45703125,
+      "reward_std": 0.1958453767001629,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.41796875,
+      "step": 77
+    },
+    {
+      "completion_length": 268.171875,
+      "epoch": 0.09056603773584905,
+      "grad_norm": 0.5879416696261314,
+      "kl": 0.108642578125,
+      "learning_rate": 9.904613311577695e-07,
+      "loss": 0.0044,
+      "reward": 0.556640625,
+      "reward_std": 0.22046153992414474,
+      "rewards/correctness_reward_func": 0.140625,
+      "rewards/strict_format_reward_func": 0.416015625,
+      "step": 78
+    },
+    {
+      "completion_length": 231.109375,
+      "epoch": 0.09172714078374455,
+      "grad_norm": 0.8976017470783989,
+      "kl": 0.18408203125,
+      "learning_rate": 9.900921597356855e-07,
+      "loss": 0.0074,
+      "reward": 0.515625,
+      "reward_std": 0.20618988201022148,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.4296875,
+      "step": 79
+    },
+    {
+      "completion_length": 196.59765625,
+      "epoch": 0.09288824383164006,
+      "grad_norm": 17.77907159035059,
+      "kl": 0.263916015625,
+      "learning_rate": 9.897160507969735e-07,
+      "loss": 0.0106,
+      "reward": 0.47265625,
+      "reward_std": 0.19209052622318268,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.41796875,
+      "step": 80
+    },
+    {
+      "completion_length": 110.4609375,
+      "epoch": 0.09404934687953556,
+      "grad_norm": 1.3276411102732406,
+      "kl": 0.3662109375,
+      "learning_rate": 9.893330096656573e-07,
+      "loss": 0.0147,
+      "reward": 0.578125,
+      "reward_std": 0.27833693847060204,
+      "rewards/correctness_reward_func": 0.140625,
+      "rewards/strict_format_reward_func": 0.4375,
+      "step": 81
+    },
+    {
+      "completion_length": 77.88671875,
+      "epoch": 0.09521044992743107,
+      "grad_norm": 1.9379785150930775,
+      "kl": 0.51953125,
+      "learning_rate": 9.889430417638883e-07,
+      "loss": 0.0208,
+      "reward": 0.51171875,
+      "reward_std": 0.1544804871082306,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.45703125,
+      "step": 82
+    },
+    {
+      "completion_length": 73.9375,
+      "epoch": 0.09637155297532655,
+      "grad_norm": 0.8473473359273695,
+      "kl": 0.48486328125,
+      "learning_rate": 9.885461526118713e-07,
+      "loss": 0.0194,
+      "reward": 0.50390625,
+      "reward_std": 0.14182795584201813,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.45703125,
+      "step": 83
+    },
+    {
+      "completion_length": 56.2109375,
+      "epoch": 0.09753265602322206,
+      "grad_norm": 2.281085454498393,
+      "kl": 0.49072265625,
+      "learning_rate": 9.88142347827784e-07,
+      "loss": 0.0196,
+      "reward": 0.4765625,
+      "reward_std": 0.12851098738610744,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.453125,
+      "step": 84
+    },
+    {
+      "completion_length": 37.875,
+      "epoch": 0.09869375907111756,
+      "grad_norm": 0.5788019582854786,
+      "kl": 0.5927734375,
+      "learning_rate": 9.877316331276993e-07,
+      "loss": 0.0237,
+      "reward": 0.470703125,
+      "reward_std": 0.055291797965765,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.470703125,
+      "step": 85
+    },
+    {
+      "completion_length": 33.50390625,
+      "epoch": 0.09985486211901307,
+      "grad_norm": 6.104259815482374,
+      "kl": 0.5439453125,
+      "learning_rate": 9.873140143255034e-07,
+      "loss": 0.0217,
+      "reward": 0.517578125,
+      "reward_std": 0.07212539482861757,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.470703125,
+      "step": 86
+    },
+    {
+      "completion_length": 20.76171875,
+      "epoch": 0.10101596516690857,
+      "grad_norm": 0.507931311814518,
+      "kl": 0.5947265625,
+      "learning_rate": 9.86889497332814e-07,
+      "loss": 0.0238,
+      "reward": 0.486328125,
+      "reward_std": 0.02734375,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 87
+    },
+    {
+      "completion_length": 16.58203125,
+      "epoch": 0.10217706821480406,
+      "grad_norm": 0.45325431756004353,
+      "kl": 0.6630859375,
+      "learning_rate": 9.864580881588958e-07,
+      "loss": 0.0265,
+      "reward": 0.501953125,
+      "reward_std": 0.024041798897087574,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.494140625,
+      "step": 88
+    },
+    {
+      "completion_length": 16.12109375,
+      "epoch": 0.10333817126269956,
+      "grad_norm": 0.38014072113379394,
+      "kl": 0.6513671875,
+      "learning_rate": 9.860197929105767e-07,
+      "loss": 0.0261,
+      "reward": 0.49609375,
+      "reward_std": 0.0078125,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.49609375,
+      "step": 89
+    },
+    {
+      "completion_length": 19.7265625,
+      "epoch": 0.10449927431059507,
+      "grad_norm": 3.246980620297076,
+      "kl": 0.6064453125,
+      "learning_rate": 9.8557461779216e-07,
+      "loss": 0.0243,
+      "reward": 0.48828125,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 90
+    },
+    {
+      "completion_length": 15.8359375,
+      "epoch": 0.10566037735849057,
+      "grad_norm": 1.125767741710113,
+      "kl": 0.6162109375,
+      "learning_rate": 9.85122569105338e-07,
+      "loss": 0.0247,
+      "reward": 0.49609375,
+      "reward_std": 0.0078125,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.49609375,
+      "step": 91
+    },
+    {
+      "completion_length": 22.44921875,
+      "epoch": 0.10682148040638606,
+      "grad_norm": 0.44874987843781083,
+      "kl": 0.6083984375,
+      "learning_rate": 9.846636532491012e-07,
+      "loss": 0.0243,
+      "reward": 0.48828125,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 92
+    },
+    {
+      "completion_length": 16.21484375,
+      "epoch": 0.10798258345428156,
+      "grad_norm": 2.7244778647827363,
+      "kl": 0.796875,
+      "learning_rate": 9.841978767196493e-07,
+      "loss": 0.0319,
+      "reward": 0.494140625,
+      "reward_std": 0.01171875,
+      "rewards/correctness_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.494140625,
+      "step": 93
+    },
+    {
+      "completion_length": 18.08984375,
+      "epoch": 0.10914368650217707,
+      "grad_norm": 0.900717061581606,
+      "kl": 0.5712890625,
+      "learning_rate": 9.83725246110298e-07,
+      "loss": 0.0229,
+      "reward": 0.49609375,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 94
+    },
+    {
+      "completion_length": 19.921875,
+      "epoch": 0.11030478955007257,
+      "grad_norm": 0.2922094758318377,
+      "kl": 0.62109375,
+      "learning_rate": 9.832457681113865e-07,
+      "loss": 0.0249,
+      "reward": 0.51953125,
+      "reward_std": 0.0546875,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 95
+    },
+    {
+      "completion_length": 22.48828125,
+      "epoch": 0.11146589259796807,
+      "grad_norm": 1.5719055280514225,
+      "kl": 0.5966796875,
+      "learning_rate": 9.827594495101822e-07,
+      "loss": 0.0239,
+      "reward": 0.53125,
+      "reward_std": 0.049292195588350296,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.4921875,
+      "step": 96
+    },
+    {
+      "completion_length": 21.078125,
+      "epoch": 0.11262699564586356,
+      "grad_norm": 1.875138112811585,
+      "kl": 0.62890625,
+      "learning_rate": 9.822662971907852e-07,
+      "loss": 0.0252,
+      "reward": 0.505859375,
+      "reward_std": 0.047479298897087574,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 97
+    },
+    {
+      "completion_length": 20.328125,
+      "epoch": 0.11378809869375907,
+      "grad_norm": 0.46440948422543626,
+      "kl": 0.5703125,
+      "learning_rate": 9.8176631813403e-07,
+      "loss": 0.0228,
+      "reward": 0.5078125,
+      "reward_std": 0.015625,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.5,
+      "step": 98
+    },
+    {
+      "completion_length": 17.6953125,
+      "epoch": 0.11494920174165457,
+      "grad_norm": 2.350666772323524,
+      "kl": 0.6298828125,
+      "learning_rate": 9.812595194173874e-07,
+      "loss": 0.0252,
+      "reward": 0.50390625,
+      "reward_std": 0.0234375,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.49609375,
+      "step": 99
+    },
+    {
+      "completion_length": 30.88671875,
+      "epoch": 0.11611030478955008,
+      "grad_norm": 0.6769899791454503,
+      "kl": 0.537109375,
+      "learning_rate": 9.807459082148648e-07,
+      "loss": 0.0215,
+      "reward": 0.56640625,
+      "reward_std": 0.0804273895919323,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 100
+    },
+    {
+      "completion_length": 38.8203125,
+      "epoch": 0.11727140783744558,
+      "grad_norm": 9.613286964921125,
+      "kl": 0.53515625,
+      "learning_rate": 9.802254917969032e-07,
+      "loss": 0.0214,
+      "reward": 0.5390625,
+      "reward_std": 0.1328125,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 101
+    },
+    {
+      "completion_length": 21.05078125,
+      "epoch": 0.11843251088534107,
+      "grad_norm": 0.383590636095333,
+      "kl": 0.5419921875,
+      "learning_rate": 9.796982775302754e-07,
+      "loss": 0.0217,
+      "reward": 0.52734375,
+      "reward_std": 0.057104695588350296,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.49609375,
+      "step": 102
+    },
+    {
+      "completion_length": 26.5234375,
+      "epoch": 0.11959361393323657,
+      "grad_norm": 0.45036244814830856,
+      "kl": 0.53759765625,
+      "learning_rate": 9.79164272877981e-07,
+      "loss": 0.0215,
+      "reward": 0.53125,
+      "reward_std": 0.090448047965765,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 103
+    },
+    {
+      "completion_length": 23.54296875,
+      "epoch": 0.12075471698113208,
+      "grad_norm": 2.2388808963947713,
+      "kl": 0.5615234375,
+      "learning_rate": 9.786234853991418e-07,
+      "loss": 0.0224,
+      "reward": 0.56640625,
+      "reward_std": 0.08606802299618721,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 104
+    },
+    {
+      "completion_length": 53.6875,
+      "epoch": 0.12191582002902758,
+      "grad_norm": 1.3962986009964624,
+      "kl": 0.515625,
+      "learning_rate": 9.780759227488937e-07,
+      "loss": 0.0207,
+      "reward": 0.55859375,
+      "reward_std": 0.1321367770433426,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 105
+    },
+    {
+      "completion_length": 45.9140625,
+      "epoch": 0.12307692307692308,
+      "grad_norm": 0.6991087377799369,
+      "kl": 0.5751953125,
+      "learning_rate": 9.775215926782787e-07,
+      "loss": 0.023,
+      "reward": 0.5390625,
+      "reward_std": 0.12271332554519176,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 106
+    },
+    {
+      "completion_length": 41.71484375,
+      "epoch": 0.12423802612481857,
+      "grad_norm": 1.113559415990131,
+      "kl": 0.501953125,
+      "learning_rate": 9.769605030341355e-07,
+      "loss": 0.0201,
+      "reward": 0.494140625,
+      "reward_std": 0.039666798897087574,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 107
+    },
+    {
+      "completion_length": 30.234375,
+      "epoch": 0.1253991291727141,
+      "grad_norm": 9.444282741370284,
+      "kl": 0.52587890625,
+      "learning_rate": 9.763926617589884e-07,
+      "loss": 0.021,
+      "reward": 0.529296875,
+      "reward_std": 0.09918868914246559,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 108
+    },
+    {
+      "completion_length": 36.328125,
+      "epoch": 0.12656023222060958,
+      "grad_norm": 0.5876435788826965,
+      "kl": 0.49365234375,
+      "learning_rate": 9.758180768909336e-07,
+      "loss": 0.0198,
+      "reward": 0.5625,
+      "reward_std": 0.1117921955883503,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.4921875,
+      "step": 109
+    },
+    {
+      "completion_length": 45.109375,
+      "epoch": 0.12772133526850507,
+      "grad_norm": 0.6967401340285807,
+      "kl": 0.58203125,
+      "learning_rate": 9.75236756563528e-07,
+      "loss": 0.0233,
+      "reward": 0.564453125,
+      "reward_std": 0.1313234455883503,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 110
+    },
+    {
+      "completion_length": 22.265625,
+      "epoch": 0.1288824383164006,
+      "grad_norm": 4.157806679866693,
+      "kl": 0.5859375,
+      "learning_rate": 9.746487090056712e-07,
+      "loss": 0.0235,
+      "reward": 0.57421875,
+      "reward_std": 0.1376468911767006,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 111
+    },
+    {
+      "completion_length": 43.33203125,
+      "epoch": 0.13004354136429608,
+      "grad_norm": 1.3608549571109825,
+      "kl": 0.52099609375,
+      "learning_rate": 9.740539425414911e-07,
+      "loss": 0.0208,
+      "reward": 0.548828125,
+      "reward_std": 0.1078859455883503,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.494140625,
+      "step": 112
+    },
+    {
+      "completion_length": 27.4609375,
+      "epoch": 0.1312046444121916,
+      "grad_norm": 0.9303195344407332,
+      "kl": 0.5302734375,
+      "learning_rate": 9.734524655902252e-07,
+      "loss": 0.0212,
+      "reward": 0.5,
+      "reward_std": 0.03125,
+      "rewards/correctness_reward_func": 0.0078125,
+      "rewards/strict_format_reward_func": 0.4921875,
+      "step": 113
+    },
+    {
+      "completion_length": 44.421875,
+      "epoch": 0.13236574746008709,
+      "grad_norm": 2.5989058419028512,
+      "kl": 0.5224609375,
+      "learning_rate": 9.728442866661012e-07,
+      "loss": 0.0209,
+      "reward": 0.5390625,
+      "reward_std": 0.09286524448543787,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 114
+    },
+    {
+      "completion_length": 55.43359375,
+      "epoch": 0.13352685050798258,
+      "grad_norm": 4.35900633763163,
+      "kl": 0.513671875,
+      "learning_rate": 9.722294143782171e-07,
+      "loss": 0.0205,
+      "reward": 0.5234375,
+      "reward_std": 0.09476527757942677,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 115
+    },
+    {
+      "completion_length": 32.14453125,
+      "epoch": 0.1346879535558781,
+      "grad_norm": 2.5576115813189664,
+      "kl": 0.48486328125,
+      "learning_rate": 9.71607857430419e-07,
+      "loss": 0.0194,
+      "reward": 0.548828125,
+      "reward_std": 0.1358339935541153,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 116
+    },
+    {
+      "completion_length": 40.48828125,
+      "epoch": 0.13584905660377358,
+      "grad_norm": 5.3055382808430025,
+      "kl": 0.55078125,
+      "learning_rate": 9.709796246211776e-07,
+      "loss": 0.022,
+      "reward": 0.599609375,
+      "reward_std": 0.1770055592060089,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 117
+    },
+    {
+      "completion_length": 48.4296875,
+      "epoch": 0.13701015965166907,
+      "grad_norm": 1.9357783363747145,
+      "kl": 0.52734375,
+      "learning_rate": 9.70344724843465e-07,
+      "loss": 0.0211,
+      "reward": 0.623046875,
+      "reward_std": 0.0844484455883503,
+      "rewards/correctness_reward_func": 0.1328125,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 118
+    },
+    {
+      "completion_length": 33.06640625,
+      "epoch": 0.1381712626995646,
+      "grad_norm": 0.9093609363691819,
+      "kl": 0.48876953125,
+      "learning_rate": 9.697031670846264e-07,
+      "loss": 0.0195,
+      "reward": 0.595703125,
+      "reward_std": 0.1571781411767006,
+      "rewards/correctness_reward_func": 0.109375,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 119
+    },
+    {
+      "completion_length": 26.15234375,
+      "epoch": 0.13933236574746008,
+      "grad_norm": 1.2956378319464605,
+      "kl": 0.52490234375,
+      "learning_rate": 9.690549604262554e-07,
+      "loss": 0.021,
+      "reward": 0.611328125,
+      "reward_std": 0.1670839935541153,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.494140625,
+      "step": 120
+    },
+    {
+      "completion_length": 23.31640625,
+      "epoch": 0.1404934687953556,
+      "grad_norm": 1.0404878413954166,
+      "kl": 0.53173828125,
+      "learning_rate": 9.684001140440638e-07,
+      "loss": 0.0213,
+      "reward": 0.533203125,
+      "reward_std": 0.08553362637758255,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 121
+    },
+    {
+      "completion_length": 33.63671875,
+      "epoch": 0.1416545718432511,
+      "grad_norm": 0.5491271196977439,
+      "kl": 0.5263671875,
+      "learning_rate": 9.677386372077523e-07,
+      "loss": 0.0211,
+      "reward": 0.537109375,
+      "reward_std": 0.0798230916261673,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 122
+    },
+    {
+      "completion_length": 27.7578125,
+      "epoch": 0.14281567489114658,
+      "grad_norm": 1.4086546997823266,
+      "kl": 0.49462890625,
+      "learning_rate": 9.670705392808795e-07,
+      "loss": 0.0198,
+      "reward": 0.642578125,
+      "reward_std": 0.14176217839121819,
+      "rewards/correctness_reward_func": 0.15625,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 123
+    },
+    {
+      "completion_length": 46.31640625,
+      "epoch": 0.1439767779390421,
+      "grad_norm": 0.44935314029572543,
+      "kl": 0.4541015625,
+      "learning_rate": 9.663958297207285e-07,
+      "loss": 0.0181,
+      "reward": 0.5390625,
+      "reward_std": 0.12601527571678162,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 124
+    },
+    {
+      "completion_length": 42.33203125,
+      "epoch": 0.14513788098693758,
+      "grad_norm": 4.134753027574554,
+      "kl": 0.45703125,
+      "learning_rate": 9.657145180781742e-07,
+      "loss": 0.0183,
+      "reward": 0.513671875,
+      "reward_std": 0.06089614052325487,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 125
+    },
+    {
+      "completion_length": 42.03515625,
+      "epoch": 0.1462989840348331,
+      "grad_norm": 0.4544547991850145,
+      "kl": 0.40625,
+      "learning_rate": 9.650266139975473e-07,
+      "loss": 0.0163,
+      "reward": 0.5703125,
+      "reward_std": 0.12411524076014757,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 126
+    },
+    {
+      "completion_length": 31.16796875,
+      "epoch": 0.1474600870827286,
+      "grad_norm": 1.6260031691388108,
+      "kl": 0.486328125,
+      "learning_rate": 9.64332127216498e-07,
+      "loss": 0.0194,
+      "reward": 0.51171875,
+      "reward_std": 0.0390625,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 127
+    },
+    {
+      "completion_length": 32.0,
+      "epoch": 0.14862119013062408,
+      "grad_norm": 0.5382173674869626,
+      "kl": 0.4267578125,
+      "learning_rate": 9.63631067565858e-07,
+      "loss": 0.0171,
+      "reward": 0.529296875,
+      "reward_std": 0.053198445588350296,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 128
+    },
+    {
+      "completion_length": 44.75390625,
+      "epoch": 0.1497822931785196,
+      "grad_norm": 0.7683808964372313,
+      "kl": 0.4228515625,
+      "learning_rate": 9.629234449695015e-07,
+      "loss": 0.0169,
+      "reward": 0.5625,
+      "reward_std": 0.1478765867650509,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 129
+    },
+    {
+      "completion_length": 40.52734375,
+      "epoch": 0.1509433962264151,
+      "grad_norm": 1.0117014376716735,
+      "kl": 0.45263671875,
+      "learning_rate": 9.62209269444205e-07,
+      "loss": 0.0181,
+      "reward": 0.65625,
+      "reward_std": 0.218528650701046,
+      "rewards/correctness_reward_func": 0.171875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 130
+    },
+    {
+      "completion_length": 28.6484375,
+      "epoch": 0.1521044992743106,
+      "grad_norm": 0.42513248914448737,
+      "kl": 0.35986328125,
+      "learning_rate": 9.614885510995047e-07,
+      "loss": 0.0144,
+      "reward": 0.5390625,
+      "reward_std": 0.10979307442903519,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 131
+    },
+    {
+      "completion_length": 48.9765625,
+      "epoch": 0.1532656023222061,
+      "grad_norm": 0.3690713785403208,
+      "kl": 0.443359375,
+      "learning_rate": 9.607613001375545e-07,
+      "loss": 0.0177,
+      "reward": 0.6015625,
+      "reward_std": 0.1454593911767006,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 132
+    },
+    {
+      "completion_length": 40.62890625,
+      "epoch": 0.15442670537010159,
+      "grad_norm": 0.3570306428369935,
+      "kl": 0.40283203125,
+      "learning_rate": 9.600275268529806e-07,
+      "loss": 0.0161,
+      "reward": 0.595703125,
+      "reward_std": 0.0766359455883503,
+      "rewards/correctness_reward_func": 0.1015625,
+      "rewards/strict_format_reward_func": 0.494140625,
+      "step": 133
+    },
+    {
+      "completion_length": 48.47265625,
+      "epoch": 0.1555878084179971,
+      "grad_norm": 0.6699794243427841,
+      "kl": 0.40283203125,
+      "learning_rate": 9.592872416327364e-07,
+      "loss": 0.0161,
+      "reward": 0.537109375,
+      "reward_std": 0.1280214972794056,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 134
+    },
+    {
+      "completion_length": 43.74609375,
+      "epoch": 0.1567489114658926,
+      "grad_norm": 0.8744900024283108,
+      "kl": 0.42919921875,
+      "learning_rate": 9.58540454955955e-07,
+      "loss": 0.0172,
+      "reward": 0.568359375,
+      "reward_std": 0.12251138500869274,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.474609375,
+      "step": 135
+    },
+    {
+      "completion_length": 35.98828125,
+      "epoch": 0.1579100145137881,
+      "grad_norm": 10.83989251533941,
+      "kl": 0.43994140625,
+      "learning_rate": 9.57787177393801e-07,
+      "loss": 0.0176,
+      "reward": 0.59765625,
+      "reward_std": 0.1689911223948002,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 136
+    },
+    {
+      "completion_length": 47.265625,
+      "epoch": 0.1590711175616836,
+      "grad_norm": 1.3945795045826723,
+      "kl": 0.4638671875,
+      "learning_rate": 9.570274196093216e-07,
+      "loss": 0.0186,
+      "reward": 0.5,
+      "reward_std": 0.051385548897087574,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 137
+    },
+    {
+      "completion_length": 37.30859375,
+      "epoch": 0.1602322206095791,
+      "grad_norm": 1.2085779669420207,
+      "kl": 0.4755859375,
+      "learning_rate": 9.562611923572943e-07,
+      "loss": 0.019,
+      "reward": 0.587890625,
+      "reward_std": 0.16475876048207283,
+      "rewards/correctness_reward_func": 0.109375,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 138
+    },
+    {
+      "completion_length": 62.33203125,
+      "epoch": 0.1613933236574746,
+      "grad_norm": 0.6440211313776459,
+      "kl": 0.419921875,
+      "learning_rate": 9.554885064840755e-07,
+      "loss": 0.0168,
+      "reward": 0.51171875,
+      "reward_std": 0.129510547965765,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.47265625,
+      "step": 139
+    },
+    {
+      "completion_length": 42.2734375,
+      "epoch": 0.1625544267053701,
+      "grad_norm": 4.184412362449491,
+      "kl": 0.458984375,
+      "learning_rate": 9.547093729274473e-07,
+      "loss": 0.0183,
+      "reward": 0.56640625,
+      "reward_std": 0.1319277435541153,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 140
+    },
+    {
+      "completion_length": 30.3203125,
+      "epoch": 0.16371552975326562,
+      "grad_norm": 0.9148362673963394,
+      "kl": 0.4169921875,
+      "learning_rate": 9.539238027164618e-07,
+      "loss": 0.0167,
+      "reward": 0.513671875,
+      "reward_std": 0.053198445588350296,
+      "rewards/correctness_reward_func": 0.0234375,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 141
+    },
+    {
+      "completion_length": 26.02734375,
+      "epoch": 0.1648766328011611,
+      "grad_norm": 23.339183218639594,
+      "kl": 0.4833984375,
+      "learning_rate": 9.531318069712853e-07,
+      "loss": 0.0193,
+      "reward": 0.599609375,
+      "reward_std": 0.15335902571678162,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 142
+    },
+    {
+      "completion_length": 39.70703125,
+      "epoch": 0.1660377358490566,
+      "grad_norm": 4.918732336995348,
+      "kl": 0.546875,
+      "learning_rate": 9.523333969030412e-07,
+      "loss": 0.0219,
+      "reward": 0.623046875,
+      "reward_std": 0.18786722049117088,
+      "rewards/correctness_reward_func": 0.1328125,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 143
+    },
+    {
+      "completion_length": 24.69140625,
+      "epoch": 0.1671988388969521,
+      "grad_norm": 2.3195586374061645,
+      "kl": 0.46875,
+      "learning_rate": 9.51528583813651e-07,
+      "loss": 0.0187,
+      "reward": 0.58203125,
+      "reward_std": 0.13684775680303574,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 144
+    },
+    {
+      "completion_length": 40.7109375,
+      "epoch": 0.1683599419448476,
+      "grad_norm": 0.6860447379166801,
+      "kl": 0.49267578125,
+      "learning_rate": 9.507173790956745e-07,
+      "loss": 0.0197,
+      "reward": 0.583984375,
+      "reward_std": 0.1592714935541153,
+      "rewards/correctness_reward_func": 0.1015625,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 145
+    },
+    {
+      "completion_length": 60.76953125,
+      "epoch": 0.16952104499274312,
+      "grad_norm": 1.208065145658311,
+      "kl": 0.55859375,
+      "learning_rate": 9.498997942321483e-07,
+      "loss": 0.0223,
+      "reward": 0.5390625,
+      "reward_std": 0.1552504375576973,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.46875,
+      "step": 146
+    },
+    {
+      "completion_length": 34.5859375,
+      "epoch": 0.1706821480406386,
+      "grad_norm": 2.412402656297087,
+      "kl": 0.5146484375,
+      "learning_rate": 9.490758407964233e-07,
+      "loss": 0.0206,
+      "reward": 0.65234375,
+      "reward_std": 0.14944713562726974,
+      "rewards/correctness_reward_func": 0.171875,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 147
+    },
+    {
+      "completion_length": 33.71875,
+      "epoch": 0.1718432510885341,
+      "grad_norm": 1.3484678762791356,
+      "kl": 0.5146484375,
+      "learning_rate": 9.482455304520012e-07,
+      "loss": 0.0206,
+      "reward": 0.50390625,
+      "reward_std": 0.033667195588350296,
+      "rewards/correctness_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 148
+    },
+    {
+      "completion_length": 43.68359375,
+      "epoch": 0.17300435413642962,
+      "grad_norm": 4.476292060280118,
+      "kl": 0.5556640625,
+      "learning_rate": 9.474088749523687e-07,
+      "loss": 0.0222,
+      "reward": 0.51953125,
+      "reward_std": 0.051385547965765,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 149
+    },
+    {
+      "completion_length": 57.0078125,
+      "epoch": 0.1741654571843251,
+      "grad_norm": 3.243977251886513,
+      "kl": 0.470703125,
+      "learning_rate": 9.465658861408323e-07,
+      "loss": 0.0188,
+      "reward": 0.5703125,
+      "reward_std": 0.1800113394856453,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 150
+    },
+    {
+      "completion_length": 48.234375,
+      "epoch": 0.1753265602322206,
+      "grad_norm": 3.965397946904913,
+      "kl": 0.51220703125,
+      "learning_rate": 9.457165759503491e-07,
+      "loss": 0.0204,
+      "reward": 0.576171875,
+      "reward_std": 0.14397033769637346,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 151
+    },
+    {
+      "completion_length": 27.34765625,
+      "epoch": 0.1764876632801161,
+      "grad_norm": 5.956710202169321,
+      "kl": 0.52734375,
+      "learning_rate": 9.448609564033593e-07,
+      "loss": 0.0211,
+      "reward": 0.521484375,
+      "reward_std": 0.13792734686285257,
+      "rewards/correctness_reward_func": 0.046875,
+      "rewards/strict_format_reward_func": 0.474609375,
+      "step": 152
+    },
+    {
+      "completion_length": 55.0234375,
+      "epoch": 0.1776487663280116,
+      "grad_norm": 1.9878758706657031,
+      "kl": 0.625,
+      "learning_rate": 9.439990396116148e-07,
+      "loss": 0.025,
+      "reward": 0.486328125,
+      "reward_std": 0.1357191875576973,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.455078125,
+      "step": 153
+    },
+    {
+      "completion_length": 47.39453125,
+      "epoch": 0.17880986937590712,
+      "grad_norm": 0.3506677329292148,
+      "kl": 0.568359375,
+      "learning_rate": 9.431308377760093e-07,
+      "loss": 0.0228,
+      "reward": 0.587890625,
+      "reward_std": 0.2043701931834221,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.470703125,
+      "step": 154
+    },
+    {
+      "completion_length": 34.31640625,
+      "epoch": 0.1799709724238026,
+      "grad_norm": 3.844767491669304,
+      "kl": 0.51953125,
+      "learning_rate": 9.422563631864037e-07,
+      "loss": 0.0208,
+      "reward": 0.513671875,
+      "reward_std": 0.1123964935541153,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.474609375,
+      "step": 155
+    },
+    {
+      "completion_length": 35.26171875,
+      "epoch": 0.1811320754716981,
+      "grad_norm": 10.543061352265745,
+      "kl": 0.66748046875,
+      "learning_rate": 9.413756282214537e-07,
+      "loss": 0.0267,
+      "reward": 0.560546875,
+      "reward_std": 0.2216724045574665,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.466796875,
+      "step": 156
+    },
+    {
+      "completion_length": 53.375,
+      "epoch": 0.18229317851959362,
+      "grad_norm": 0.6406847195030275,
+      "kl": 0.51416015625,
+      "learning_rate": 9.404886453484335e-07,
+      "loss": 0.0206,
+      "reward": 0.486328125,
+      "reward_std": 0.12331762351095676,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.455078125,
+      "step": 157
+    },
+    {
+      "completion_length": 33.31640625,
+      "epoch": 0.1834542815674891,
+      "grad_norm": 1.254136142861394,
+      "kl": 0.6044921875,
+      "learning_rate": 9.395954271230603e-07,
+      "loss": 0.0242,
+      "reward": 0.55078125,
+      "reward_std": 0.1352296955883503,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 158
+    },
+    {
+      "completion_length": 38.45703125,
+      "epoch": 0.18461538461538463,
+      "grad_norm": 1.1951784005374884,
+      "kl": 0.580078125,
+      "learning_rate": 9.386959861893157e-07,
+      "loss": 0.0232,
+      "reward": 0.5,
+      "reward_std": 0.09967818111181259,
+      "rewards/correctness_reward_func": 0.03125,
+      "rewards/strict_format_reward_func": 0.46875,
+      "step": 159
+    },
+    {
+      "completion_length": 49.9921875,
+      "epoch": 0.18577648766328012,
+      "grad_norm": 1.0186965999705826,
+      "kl": 0.611328125,
+      "learning_rate": 9.377903352792671e-07,
+      "loss": 0.0244,
+      "reward": 0.599609375,
+      "reward_std": 0.19500285387039185,
+      "rewards/correctness_reward_func": 0.1328125,
+      "rewards/strict_format_reward_func": 0.466796875,
+      "step": 160
+    },
+    {
+      "completion_length": 25.6953125,
+      "epoch": 0.1869375907111756,
+      "grad_norm": 0.9553165570763839,
+      "kl": 0.6337890625,
+      "learning_rate": 9.368784872128876e-07,
+      "loss": 0.0253,
+      "reward": 0.65625,
+      "reward_std": 0.23248206824064255,
+      "rewards/correctness_reward_func": 0.1796875,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 161
+    },
+    {
+      "completion_length": 29.96484375,
+      "epoch": 0.18809869375907112,
+      "grad_norm": 1.320529298833274,
+      "kl": 0.677734375,
+      "learning_rate": 9.359604548978741e-07,
+      "loss": 0.0271,
+      "reward": 0.546875,
+      "reward_std": 0.1555742807686329,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 162
+    },
+    {
+      "completion_length": 37.44921875,
+      "epoch": 0.1892597968069666,
+      "grad_norm": 2.2124875651885754,
+      "kl": 0.6181640625,
+      "learning_rate": 9.35036251329465e-07,
+      "loss": 0.0247,
+      "reward": 0.513671875,
+      "reward_std": 0.09588673710823059,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.474609375,
+      "step": 163
+    },
+    {
+      "completion_length": 30.9609375,
+      "epoch": 0.19042089985486213,
+      "grad_norm": 0.7762326728671275,
+      "kl": 0.755859375,
+      "learning_rate": 9.341058895902562e-07,
+      "loss": 0.0303,
+      "reward": 0.615234375,
+      "reward_std": 0.18324187397956848,
+      "rewards/correctness_reward_func": 0.1328125,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 164
+    },
+    {
+      "completion_length": 39.90234375,
+      "epoch": 0.19158200290275762,
+      "grad_norm": 1.5333968229845967,
+      "kl": 0.681640625,
+      "learning_rate": 9.331693828500158e-07,
+      "loss": 0.0273,
+      "reward": 0.53125,
+      "reward_std": 0.11541798710823059,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 165
+    },
+    {
+      "completion_length": 39.2265625,
+      "epoch": 0.1927431059506531,
+      "grad_norm": 3.341316216605523,
+      "kl": 0.6484375,
+      "learning_rate": 9.322267443654972e-07,
+      "loss": 0.026,
+      "reward": 0.552734375,
+      "reward_std": 0.13462539482861757,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.474609375,
+      "step": 166
+    },
+    {
+      "completion_length": 41.33203125,
+      "epoch": 0.19390420899854863,
+      "grad_norm": 0.8334577080111663,
+      "kl": 0.689453125,
+      "learning_rate": 9.312779874802525e-07,
+      "loss": 0.0276,
+      "reward": 0.72265625,
+      "reward_std": 0.2527032420039177,
+      "rewards/correctness_reward_func": 0.2578125,
+      "rewards/strict_format_reward_func": 0.46484375,
+      "step": 167
+    },
+    {
+      "completion_length": 27.4296875,
+      "epoch": 0.19506531204644412,
+      "grad_norm": 0.6225355591311358,
+      "kl": 0.6796875,
+      "learning_rate": 9.303231256244428e-07,
+      "loss": 0.0272,
+      "reward": 0.572265625,
+      "reward_std": 0.1169070415198803,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 168
+    },
+    {
+      "completion_length": 26.65625,
+      "epoch": 0.19622641509433963,
+      "grad_norm": 0.31164342343966356,
+      "kl": 0.701171875,
+      "learning_rate": 9.293621723146484e-07,
+      "loss": 0.0281,
+      "reward": 0.595703125,
+      "reward_std": 0.1571781374514103,
+      "rewards/correctness_reward_func": 0.109375,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 169
+    },
+    {
+      "completion_length": 32.9609375,
+      "epoch": 0.19738751814223512,
+      "grad_norm": 0.8220338043159918,
+      "kl": 0.6875,
+      "learning_rate": 9.283951411536772e-07,
+      "loss": 0.0274,
+      "reward": 0.59375,
+      "reward_std": 0.15472763776779175,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 170
+    },
+    {
+      "completion_length": 29.8984375,
+      "epoch": 0.19854862119013061,
+      "grad_norm": 3.2547872245531146,
+      "kl": 0.7255859375,
+      "learning_rate": 9.274220458303726e-07,
+      "loss": 0.029,
+      "reward": 0.609375,
+      "reward_std": 0.2199743092060089,
+      "rewards/correctness_reward_func": 0.1328125,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 171
+    },
+    {
+      "completion_length": 22.08203125,
+      "epoch": 0.19970972423802613,
+      "grad_norm": 1.0413843485165826,
+      "kl": 0.7294921875,
+      "learning_rate": 9.264429001194192e-07,
+      "loss": 0.0292,
+      "reward": 0.595703125,
+      "reward_std": 0.2031683847308159,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 172
+    },
+    {
+      "completion_length": 51.3671875,
+      "epoch": 0.20087082728592162,
+      "grad_norm": 3.778010417040405,
+      "kl": 0.6826171875,
+      "learning_rate": 9.254577178811482e-07,
+      "loss": 0.0273,
+      "reward": 0.6015625,
+      "reward_std": 0.1914496347308159,
+      "rewards/correctness_reward_func": 0.1328125,
+      "rewards/strict_format_reward_func": 0.46875,
+      "step": 173
+    },
+    {
+      "completion_length": 44.84765625,
+      "epoch": 0.20203193033381714,
+      "grad_norm": 2.305304587495654,
+      "kl": 0.7666015625,
+      "learning_rate": 9.24466513061341e-07,
+      "loss": 0.0307,
+      "reward": 0.59375,
+      "reward_std": 0.1900114268064499,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 174
+    },
+    {
+      "completion_length": 36.86328125,
+      "epoch": 0.20319303338171263,
+      "grad_norm": 0.7728198355409298,
+      "kl": 0.712890625,
+      "learning_rate": 9.234692996910323e-07,
+      "loss": 0.0285,
+      "reward": 0.62890625,
+      "reward_std": 0.1866152435541153,
+      "rewards/correctness_reward_func": 0.1484375,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 175
+    },
+    {
+      "completion_length": 29.98828125,
+      "epoch": 0.20435413642960812,
+      "grad_norm": 1.289972465399101,
+      "kl": 0.6865234375,
+      "learning_rate": 9.224660918863103e-07,
+      "loss": 0.0275,
+      "reward": 0.595703125,
+      "reward_std": 0.13913594186306,
+      "rewards/correctness_reward_func": 0.109375,
+      "rewards/strict_format_reward_func": 0.486328125,
+      "step": 176
+    },
+    {
+      "completion_length": 34.23046875,
+      "epoch": 0.20551523947750364,
+      "grad_norm": 0.5793735600612212,
+      "kl": 0.759765625,
+      "learning_rate": 9.214569038481181e-07,
+      "loss": 0.0304,
+      "reward": 0.5625,
+      "reward_std": 0.1687820851802826,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 177
+    },
+    {
+      "completion_length": 31.23828125,
+      "epoch": 0.20667634252539913,
+      "grad_norm": 0.5984526162665614,
+      "kl": 0.771484375,
+      "learning_rate": 9.20441749862052e-07,
+      "loss": 0.0309,
+      "reward": 0.68359375,
+      "reward_std": 0.1762864589691162,
+      "rewards/correctness_reward_func": 0.1953125,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 178
+    },
+    {
+      "completion_length": 38.296875,
+      "epoch": 0.20783744557329464,
+      "grad_norm": 1.181839300217833,
+      "kl": 0.7099609375,
+      "learning_rate": 9.1942064429816e-07,
+      "loss": 0.0284,
+      "reward": 0.548828125,
+      "reward_std": 0.10820978879928589,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.470703125,
+      "step": 179
+    },
+    {
+      "completion_length": 35.44921875,
+      "epoch": 0.20899854862119013,
+      "grad_norm": 7.06190367002311,
+      "kl": 0.7001953125,
+      "learning_rate": 9.18393601610737e-07,
+      "loss": 0.028,
+      "reward": 0.56640625,
+      "reward_std": 0.1620996668934822,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 180
+    },
+    {
+      "completion_length": 24.76953125,
+      "epoch": 0.21015965166908562,
+      "grad_norm": 0.4282018680025159,
+      "kl": 0.7275390625,
+      "learning_rate": 9.173606363381218e-07,
+      "loss": 0.0291,
+      "reward": 0.654296875,
+      "reward_std": 0.2599649503827095,
+      "rewards/correctness_reward_func": 0.1640625,
+      "rewards/strict_format_reward_func": 0.490234375,
+      "step": 181
+    },
+    {
+      "completion_length": 61.73046875,
+      "epoch": 0.21132075471698114,
+      "grad_norm": 0.666578405400966,
+      "kl": 0.705078125,
+      "learning_rate": 9.1632176310249e-07,
+      "loss": 0.0283,
+      "reward": 0.54296875,
+      "reward_std": 0.1953125,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.46484375,
+      "step": 182
+    },
+    {
+      "completion_length": 41.6875,
+      "epoch": 0.21248185776487663,
+      "grad_norm": 2.4086312779043495,
+      "kl": 0.6748046875,
+      "learning_rate": 9.152769966096481e-07,
+      "loss": 0.027,
+      "reward": 0.65625,
+      "reward_std": 0.2158661186695099,
+      "rewards/correctness_reward_func": 0.1796875,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 183
+    },
+    {
+      "completion_length": 47.5078125,
+      "epoch": 0.21364296081277212,
+      "grad_norm": 0.7460531004888707,
+      "kl": 0.6865234375,
+      "learning_rate": 9.142263516488244e-07,
+      "loss": 0.0275,
+      "reward": 0.53125,
+      "reward_std": 0.1276262328028679,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 184
+    },
+    {
+      "completion_length": 38.33203125,
+      "epoch": 0.21480406386066764,
+      "grad_norm": 1.677852493854183,
+      "kl": 0.6611328125,
+      "learning_rate": 9.131698430924605e-07,
+      "loss": 0.0264,
+      "reward": 0.6015625,
+      "reward_std": 0.1689911223948002,
+      "rewards/correctness_reward_func": 0.1171875,
+      "rewards/strict_format_reward_func": 0.484375,
+      "step": 185
+    },
+    {
+      "completion_length": 31.5859375,
+      "epoch": 0.21596516690856313,
+      "grad_norm": 21.23887712764663,
+      "kl": 0.689453125,
+      "learning_rate": 9.121074858959996e-07,
+      "loss": 0.0275,
+      "reward": 0.564453125,
+      "reward_std": 0.09907388500869274,
+      "rewards/correctness_reward_func": 0.0859375,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 186
+    },
+    {
+      "completion_length": 41.9453125,
+      "epoch": 0.21712626995645865,
+      "grad_norm": 0.6508903252265714,
+      "kl": 0.6748046875,
+      "learning_rate": 9.110392950976764e-07,
+      "loss": 0.027,
+      "reward": 0.5234375,
+      "reward_std": 0.137323047965765,
+      "rewards/correctness_reward_func": 0.0546875,
+      "rewards/strict_format_reward_func": 0.46875,
+      "step": 187
+    },
+    {
+      "completion_length": 62.2109375,
+      "epoch": 0.21828737300435413,
+      "grad_norm": 1.1278006956369315,
+      "kl": 0.673828125,
+      "learning_rate": 9.099652858183027e-07,
+      "loss": 0.027,
+      "reward": 0.56640625,
+      "reward_std": 0.1318129375576973,
+      "rewards/correctness_reward_func": 0.1015625,
+      "rewards/strict_format_reward_func": 0.46484375,
+      "step": 188
+    },
+    {
+      "completion_length": 46.51171875,
+      "epoch": 0.21944847605224962,
+      "grad_norm": 1.1019281044279183,
+      "kl": 0.7548828125,
+      "learning_rate": 9.088854732610542e-07,
+      "loss": 0.0302,
+      "reward": 0.572265625,
+      "reward_std": 0.1169070415198803,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 189
+    },
+    {
+      "completion_length": 35.03125,
+      "epoch": 0.22060957910014514,
+      "grad_norm": 14.962735092929739,
+      "kl": 0.6845703125,
+      "learning_rate": 9.077998727112552e-07,
+      "loss": 0.0274,
+      "reward": 0.564453125,
+      "reward_std": 0.1369277872145176,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.470703125,
+      "step": 190
+    },
+    {
+      "completion_length": 55.27734375,
+      "epoch": 0.22177068214804063,
+      "grad_norm": 0.3504770063047999,
+      "kl": 0.654296875,
+      "learning_rate": 9.067084995361623e-07,
+      "loss": 0.0262,
+      "reward": 0.55078125,
+      "reward_std": 0.1510637328028679,
+      "rewards/correctness_reward_func": 0.078125,
+      "rewards/strict_format_reward_func": 0.47265625,
+      "step": 191
+    },
+    {
+      "completion_length": 47.34375,
+      "epoch": 0.22293178519593615,
+      "grad_norm": 1.8159467800873224,
+      "kl": 0.7333984375,
+      "learning_rate": 9.05611369184746e-07,
+      "loss": 0.0293,
+      "reward": 0.57421875,
+      "reward_std": 0.1376468911767006,
+      "rewards/correctness_reward_func": 0.09375,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 192
+    },
+    {
+      "completion_length": 49.0390625,
+      "epoch": 0.22409288824383164,
+      "grad_norm": 3.120448246962212,
+      "kl": 0.69140625,
+      "learning_rate": 9.045084971874737e-07,
+      "loss": 0.0277,
+      "reward": 0.609375,
+      "reward_std": 0.1610843911767006,
+      "rewards/correctness_reward_func": 0.1328125,
+      "rewards/strict_format_reward_func": 0.4765625,
+      "step": 193
+    },
+    {
+      "completion_length": 33.7421875,
+      "epoch": 0.22525399129172713,
+      "grad_norm": 2.5542170019130244,
+      "kl": 0.6787109375,
+      "learning_rate": 9.03399899156088e-07,
+      "loss": 0.0272,
+      "reward": 0.58984375,
+      "reward_std": 0.1015625,
+      "rewards/correctness_reward_func": 0.1015625,
+      "rewards/strict_format_reward_func": 0.48828125,
+      "step": 194
+    },
+    {
+      "completion_length": 54.81640625,
+      "epoch": 0.22641509433962265,
+      "grad_norm": 0.6406888848138048,
+      "kl": 0.7216796875,
+      "learning_rate": 9.022855907833871e-07,
+      "loss": 0.0288,
+      "reward": 0.5390625,
+      "reward_std": 0.1084902435541153,
+      "rewards/correctness_reward_func": 0.0703125,
+      "rewards/strict_format_reward_func": 0.46875,
+      "step": 195
+    },
+    {
+      "completion_length": 33.41796875,
+      "epoch": 0.22757619738751814,
+      "grad_norm": 0.3997318900990298,
+      "kl": 0.7177734375,
+      "learning_rate": 9.011655878430018e-07,
+      "loss": 0.0287,
+      "reward": 0.587890625,
+      "reward_std": 0.16475242003798485,
+      "rewards/correctness_reward_func": 0.109375,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 196
+    },
+    {
+      "completion_length": 62.3671875,
+      "epoch": 0.22873730043541365,
+      "grad_norm": 1.585750352753696,
+      "kl": 0.71484375,
+      "learning_rate": 9.000399061891728e-07,
+      "loss": 0.0286,
+      "reward": 0.630859375,
+      "reward_std": 0.141229297965765,
+      "rewards/correctness_reward_func": 0.1484375,
+      "rewards/strict_format_reward_func": 0.482421875,
+      "step": 197
+    },
+    {
+      "completion_length": 38.67578125,
+      "epoch": 0.22989840348330914,
+      "grad_norm": 5.358536856283286,
+      "kl": 0.6640625,
+      "learning_rate": 8.98908561756526e-07,
+      "loss": 0.0265,
+      "reward": 0.517578125,
+      "reward_std": 0.055291798897087574,
+      "rewards/correctness_reward_func": 0.0390625,
+      "rewards/strict_format_reward_func": 0.478515625,
+      "step": 198
+    },
+    {
+      "completion_length": 30.2109375,
+      "epoch": 0.23105950653120463,
+      "grad_norm": 0.39601285118928503,
+      "kl": 0.720703125,
+      "learning_rate": 8.977715705598467e-07,
+      "loss": 0.0288,
+      "reward": 0.54296875,
+      "reward_std": 0.09165100753307343,
+      "rewards/correctness_reward_func": 0.0625,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 199
+    },
+    {
+      "completion_length": 29.21484375,
+      "epoch": 0.23222060957910015,
+      "grad_norm": 1.491626514094341,
+      "kl": 0.73046875,
+      "learning_rate": 8.966289486938536e-07,
+      "loss": 0.0292,
+      "reward": 0.63671875,
+      "reward_std": 0.1542871668934822,
+      "rewards/correctness_reward_func": 0.15625,
+      "rewards/strict_format_reward_func": 0.48046875,
+      "step": 200
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 861,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3ef0b57332a5361e40cb4439506d8b573d5a0adf1890f8b03706312320d537d
+size 6904

checkpoint-300/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-300/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-1.5B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.3",
+  "use_cache": true,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

checkpoint-300/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.48.3"
+}

checkpoint-300/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5951029c332b3867c61088e4f0129b89b3e9e89a72bf73f4294d42e7fac180f6
+size 3087467144

checkpoint-300/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>"
+}

checkpoint-300/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

checkpoint-300/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-300/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3ef0b57332a5361e40cb4439506d8b573d5a0adf1890f8b03706312320d537d
+size 6904

checkpoint-400/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-400/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-1.5B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.3",
+  "use_cache": true,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

checkpoint-400/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.48.3"
+}

checkpoint-400/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12c998a01058047085420410ae15d439dcd35ceb768c98ae5ea2e3c26f9f5647
+size 3087467144

checkpoint-400/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>"
+}

checkpoint-400/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

checkpoint-400/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-400/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3ef0b57332a5361e40cb4439506d8b573d5a0adf1890f8b03706312320d537d
+size 6904

checkpoint-500/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-500/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-1.5B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.3",
+  "use_cache": true,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

checkpoint-500/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.48.3"
+}

checkpoint-500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4101ba87a7d2a4729ffaf4253e7292528b6494b2c920f78e0972ba2549176fe9
+size 3087467144